factors affecting reading comprehension research paper

ORIGINAL RESEARCH article

Factors influencing secondary school students’ reading literacy: an analysis based on xgboost and shap methods.

• 1 Collaborative Innovation Centre for Assessment of Basic Education Quality, Beijing Normal University, Beijing, China
• 2 Institute of Education, University of Alberta, Edmonton, AB, Canada

This paper constructs a predictive model of student reading literacy based on data from students who participated in the Program for International Student Assessment (PISA 2018) from four provinces/municipalities of China, i.e., Beijing, Shanghai, Jiangsu and Zhejiang. We calculated the contribution of influencing factors in the model by using eXtreme Gradient Boosting (XGBoost) algorithm and sHapley additive exPlanations (SHAP) values, and get the following findings: (1) Factors that have the greatest impact on students’ reading literacy are from individual and family levels, with school-level factors taking a relative back seat. (2) The most important influencing factors at individual level are reading metacognition and reading interest. (3) The most important factors at family level are ESCS (index of economic, social and cultural status) and language environment, and dialect is negative for reading literacy, whereas proficiency in both a dialect and Mandarin plays a positive role. (4) At the school level, the most important factors are time dedicated to learning and class discipline, and we found that there is an optimal value for learning time, which suggests that reasonable learning time is beneficial, but overextended learning time may make academic performance worse instead of improving it.

Introduction

Reading literacy is important for gaining knowledge and understanding the world, and it is a prerequisite for individual to become a good reader ( Dreher and Mikulecky., 2000 ). The Program for International Student Assessment 2018 (PISA 2018) defined reading literacy as “understanding, using, evaluating, reflecting on and engaging with texts in order to achieve one’s goals, to develop one’s knowledge and potential and to participate in society.” PISA 2018 takes reading literacy as a foundation for full participation in contemporary society, requiring students to be able to integrate and put into practice textual information with prior knowledge while weighing the accuracy of arguments in and reflecting on the information conveyed by the text ( OECD, 2019 ). As seen in the PISA definition, today’s reading literacy is no longer a skill acquired only in the early years of education but an evolving skill and strategy, and it’s focus is no longer on collection and memorization but on acquisition and use of information ( OECD, 2010 ).

The results of the OECD Adult Skills Survey showed that literacy, numeracy and problem-solving skills are key information-processing skills for workers in the 21st century, and the survey found that workers who can make complex inferences and evaluate text claims and arguments are able to earn higher salaries than other workers, while workers with lower literacy skills face a higher risk of unemployment ( OECD, 2013 ). This suggests that reading literacy has become a prerequisite for individuals to successfully participate in life and work. In addition to being important to workers themselves, reading literacy is also crucial for enhancing a nation’s cultural soft power and competitiveness and is an important indicator of a nation’s social civilization and comprehensive national power ( Luo et al., 2016 ). Identifying the factors that significantly influence reading literacy and understanding factors influencing reading literacy would help students improve their reading literacy, which are crucial to the development of education policy making, top-level curriculum design, and improvement in classroom teaching strategies.

Students’ reading literacy is influenced by a variety of factors, including learning strategies, motivation, family support, school instruction, etc. These factors can be divided basically into three levels, i.e., individual level, family level, and school level.

Factors at individual level

The impact of individual-level factors on reading literacy has been an important topic of research because its influence on student’s academic performance is direct. Existing research on students’ individual factors mostly centered on innate factors (e.g., intelligence, gender), reading strategies (e.g., learning strategies, metacognitive strategies), and motivation (e.g., interest in reading, competitive environment).

Innate factors

Deary et al. (2007) found a high correlation between intelligence traits and academic achievement for all subjects, including reading. Lechner et al. (2019) found that fluid intelligence positively predicted initial levels of reading ability and competence in later days. Smith et al. (2012) noted that girls’ reading achievement and enjoyment were significantly higher than that of boys’, implying a gender difference in reading literacy that cannot be ignored. By analyzing PISA 2018 data, Chunjin (2020) , similarly, found significant gender differences in students’ reading literacy performance in four provinces/municipalities in China, but the effects were not as significant as other factors, while Logan (2008) noted small gender differences in reading ability but large gender differences in attention to and attitudes toward reading, and reading frequency. It has been claimed that gender affects students’ academic self-concept, motivation, and cognitive strategies ( Swalander and Taube, 2007 ). Although both reading literacy and other factors showed significant gender differences, it has been suggested that stereotypes are one of source of gender differences due to the presence of stereotypes, as teachers usually have greater academic expectations of girls ( Muntoni and Retelsdorf, 2018 ), which undermine boys’ self-concept of reading ( Retelsdorf et al., 2015 ).

Reading strategies

Ülle et al. (2015) argued that learning strategies are very closely related to reading level and that students’ use of learning strategies can effectively explain the differences on reading literacy tests. Ghafournia and Afghari (2013) found that students at a high level of reading proficiency would use learning strategies more efficiently and stated that learning strategies can act as mediators to link linguistic and nonlinguistic variables, i.e., nonlinguistic variables indirectly affect linguistic variables through learning strategies. Reading metacognitive strategy measured the level of students’ perceptions of effective reading strategies ( Jing, 2012 ), and the ability to use metacognitive strategies is closely related to students’ literacy performance ( Paris and Oka, 1986 ; Lee and Shute, 2010 ; Areepattamannil and Caleon, 2013 ). In an experimental study, Paris and Oka (1986) found that increasing students’ metacognitive knowledge of reading can increase their use of reading strategies. Similarly, Jing (2012) analyzed data from Shanghai PISA 2009 and concluded that metacognitive strategies had a highly significant effect on Shanghai students’ reading performance.

Reading motivation

Fengning et al. (2000) found that there is a highly significant positive correlation between the level of secondary school students’ reading motivation and reading performance. Retelsdorf et al. (2011) divided reading motivation into intrinsic (e.g., reading enjoyment, reading interest) and extrinsic (e.g., competition) and found that reading enjoyment had a positive effect on initial reading performance but did not affect its further improvement, whereas interest in reading was not related to initial reading level but significantly affected the improvement of reading level, whereas competition has a negative effect on reading performance but does not affect its improvement, which is consistent with the findings in the research of Unrau and Schlackman (2006) . Logan et al. (2011) argued that intrinsic motivation explains differences in the improvement of reading skills among students with low reading proficiency. And using quantile regression, Chunjin (2020) , similarly, found that interest in reading has a greater marginal effect on reading literacy for students at the lower quantile. In addition, Pokay and Blumenfeld (1990) found that the pattern of motivation that affects academic performance by influencing learning strategies differed between early and late stages of learning.

Family factors

A growing body of research has demonstrated that family factors play a very important role in the development of students’ reading skills ( Halle et al., 1997 ; Crosnoe et al., 2010 ; Hongbo et al., 2016 ); it even suggests that family engagement is a better predictor of student achievement compared to school engagement ( Tatlisu et al., 2011 ), and that schools cannot compensate for differences in reading skills because of family differences ( Banerjee and Lamb, 2016 ). Studies on family factors influencing students’ reading literacy center mainly on family financial situation and parents’ education level.

Family financial situation

It has been found that students’ academic performance is significantly and positively correlated with the family financial situation. In China, Han (2017) found that family income has a significant influence on children’s education level, and the increasing family income can improve their education level. Family influence on reading literacy exists from early childhood. For instance, Crosnoe et al. (2010) claimed that children from families of high socioeconomic status are more likely to be exposed to stimulating environment that are critical to children’s reading, and early learning differences usually exist as the child grows, and the differences in reading skills are more significant across financial levels in the later stages of learning ( Welch, 2013 ; Goldfeld et al., 2021 ). In addition, Hongbo et al. (2016) showed that families with better financial conditions can provide their children with access to more reading resources and educational chance to promote reading competence, which is consistent with findings of Silin et al. (2014) , who indicated that high-ESCS families usually possess more cultural capital. For low-ESCS families, in contrast, there are many mediating variables associated with poor reading development such as higher rates of absenteeism and mobility, and less parental encouragement of academic pursuits ( Buckingham et al., 2013 ).

Parents’ educational level

Parental educational level affect student’s ability indirectly through mediating variables such as behaviors ( Davis-Kean and Pamela, 2005 ). From the Greek census data, Davis-Kean and Pamela (2005) found that the educational achievement of daughters in the last 30 years depends significantly on the educational level of parents (especially mothers); Janet and Enrico (2003) also found a significant positive effect of parents’ educational level on children’s educational participation and academic achievement. A number of studies have found that parental education requires some mediating variables to play a role. For example, Mauldin et al. (2001) found that parental education influences the economic investment that families put into their children’s education; this is similar to the conclusion of Spagat (2006) , who argued that students with well-educated parents often take advantage of their family background and invest heavily in their human capital. In addition, there is a strong positive relationship between parents’ educational level and time spent on their children ( Guryan et al., 2008 ); moreover, parents’ educational level affects family environment as well as parent-child interaction ( Davis-Kean and Pamela, 2005 ), which in turn indirectly influence children’s academic performance via parental involvement ( Xiaoying, 2022 ).

School factors

School is the primary place where students live and learn in their daily education. School-level factors that affect students’ reading literacy are mainly school climate and type, teachers’ classroom teaching, among others.

School climate

Perparim (2014) stated that school type, school socioeconomic status, and classroom environment were significant predictors of reading performance, as well as differences in the effects of gender and family socioeconomic status on reading performance, and students in poorer schools may not have the material conditions to satisfy their reading achievement demands ( Hart et al., 2013 ). However, Bo et al. (2017) found that school size and material resources had little effect on student performance, but friendly school climate and classroom order could significantly and positively influence student performance. So school climate is also a focus of the research in this regard. Berkowitz (2021) pointed out that schools’ positive social climate could moderate the strength of the association between SES and achievement, and narrow the literacy achievement gap among students under different financial situation. Meanwhile, school climate perceptions can compensate for the negative contribution of disadvantaged family factors on academic achievement ( O’Malley et al., 2015 ).

Teacher instruction

As direct guides of students in the learning process, teachers can directly and profoundly influence students’ learning; however, this influence is exercised indirectly through mediating variables that affect students’ academic performance. For instance, teacher’s support can significantly affect students’ academic emotions. Positive emotions were regarded as a mediation variable between teacher support and academic engagement, and teacher’s support could promote positive emotions and mitigate negative ones, resulting in students’ improved academic engagement and enjoyment ( Ahmed et al., 2010 ; Ms and Syh, 2021 ). Moreover, it has been found that teacher support motivates students to read and thus enhances their reading competence, because students’ interest and attention will grow if they perceive teacher’s support ( Law, 2011 ). In addition, teachers’ perceived school environment (e.g., student support, collegiality, resource adequacy, work pressure) influenced teaching styles ( Webster and Fisher, 2003 ), and teachers’ utilization of appropriate teaching styles, such as providing students with effective and challenging learning tasks and using motivational teaching strategies to stimulate students’ interest in reading, can lead to improved student’s performance in reading comprehension ( Law, 2011 ).

Research questions

We expand our previous research by exploring a large number of possible student-level predictors gathered during the Program for International Student Assessment (PISA) by utilizing machine learning method to predict reading ability. The machine learning method employed here has notable advantages over traditional statistical analyses, especially for large datasets like PISA, for it is obviously helpful in handling complex interactions among predictors and nonlinear trend, and the regularization methods can help prevent over-fitting of models. The strength of machine learning models have been proved in the study of education tasks such as predicting dropout in MOOCs ( Xing, 2019 ), recommendation-based mobile personalized learning ( Hsu et al., 2013 ) and analyzing student’s emotion and behavior ( Ninaus et al., 2019 ). On the other hand, however, due to a lack of interpretability, researchers have to combine machine learning with interpretation algorithms in exploring the relationship between predictors and outcomes. For example, ( Bosch, 2021 ) used Shapley values to calculate the importance of each variable to mindset interventions, ( Hu et al., 2022 ) utilized Support Vector Machine—Recursive Feature Elimination (SVM-RFE) in classifying the contextual factors influencing reading performance.

In this study, we used the PISA dataset to explore two research questions, which, in addition to having confirmed previous conclusions, reveals some new interesting findings. The questions are:

RQ1: Among individual, family and school factors, which has the greatest impact on reading literacy? Besides providing an answer to this question, we also identify some variables worthy of further discussion.

RQ2: What’s the contribution of each of these variables in predicting reading literacy? In this analysis we have explored whether these variables are monotonic or linear with, and, if not, whether there is an optimal value.

Materials and methods

Data source.

The data used in this study come from the Program for International Student Assessment (PISA 2018), conducted globally by the Organization for Economic Cooperation and Development (OECD) in 2018, which evaluates whether, at the end of compulsory education, students had the skills and knowledge necessary to participate in the society of the future, rather than focusing only on whether students had mastered specific subjects. PISA 2018 used two-stage stratified sampling to draw the sample. Schools were selected using unequal probability sampling proportional to school size in the first stage, and students were randomly selected from the selected schools to participate in the assessment program in the second stage. In China, a total of 361 secondary schools and 12,058 students from Beijing, Shanghai, Jiangsu and Zhejiang participated in this program.

Data preprocessing

The 12,058 students (992,302 after weighting) from the above mentioned four provinces/municipalities of China are the objects of our analysis. Since both original questionnaire variables and composite variables were included in the dataset, the above duplicate original questionnaire variables were excluded, some variables that were not measured in the four provinces/municipalities were deleted, and some composite variables were screened, such as ESCS (index of economic, social and cultural status), which was synthesized from highest parental occupation, parental education and home possessions; and learning time, which was obtained by multiplying the weekly class hours by the average length of each class. Nearest-neighbor averaging was applyed by levels to impute missing values of continuous variables, with 134 (10174.7 after weighting) or 1.1% (1.0% after weighting) of the total samples, and we removed samples with missing categorical variables. Finally, a total of 11,924 (982,127 after weighting) samples were retained.

In the PISA data, the cognitive data are scaled with the Rasch Model and the performance of students is denoted with plausible values (PVs) ( OECD, 2009a ). PISA 2018 provided 10 plausible values (PV values), which are randomly drawn from the student’s trait distribution, to estimate the probability distribution of students’ reading literacy. Statistical analyses usually should be performed independently on each of these 10 plausible values and aggregate results, but, with the large number of samples, using one plausible value or all plausible values does not make any substantial difference ( OECD, 2009b ). Therefore, the first plausible value (PV1), which was considered as an estimate of students’ reading literacy, was introduced into the statistical model as a predicted variable.

Model training

Most studies on reading factors used structural equation models ( Halle et al., 1997 ; Sheng et al., 2014 ; Hongbo et al., 2016 ; Kaili and Wei, 2021 ), multilevel regression ( Pingping et al., 2011 ; Bo et al., 2017 ) and quantile regression ( Chunjin, 2020 ). These methods have many shortcomings. For instance, they are all linear models, which cannot identify the nonlinear relationships and handle complex interaction effect; and they have some strong assumptions such as samples independence and variables independence, which unfortunately are often not correct in reality.

Machine learning methods can provide may inspirational ideas for educational researchers because they require no specific assumption and can explore patterns contained in complex, massive data in a data-driven manner. The application of machine learning in educational task research has become increasingly popular in recent years.

This study used three machine learning models, i.e., Support Vector Regression (SVR), Random Forest Regression (RFR), and eXtreme Gradient Boosting (XGBoost), to predict students’ reading literacy. SVR is a type of linear regression, but a slack variable ∈ is added to the calculation of the loss function, and only sample points that fall outside the interval band of width 2 ∈ are counted. RFR and XGBoost are two classical models of ensemble models, which consist of multiple weak learners. RFR is based on the idea of bagging, generating several regression trees, with each constructed by some randomly selected samples and features. All weak learners learn and make predictions independently, and finally, the predictions of all weak learners are averaged to obtain the final prediction results. XGBoost is a machine learning algorithm proposed by Chen and Guestrin (2016) , which has gained wide attention in recent years due to its significant prediction accuracy. XGBoost uses a second-order Taylor series to approximate the value of the loss function and further reduces the possibility of overfitting by adding a regularization term. Besides, overfitting can be further prevented by adjusting parameters such as “maximum tree depth” and “smallest subtree weight” to give the best prediction results in the test set.

10-fold cross-validation was used in this study to adjust the hyperparameters based on the decidable coefficients R 2 on the test set, and finally the combination of hyperparameters that performed best on the test set was selected, with the model showing highest predictive ability being chosen.

Model interpretation

The complexity of the models of machine learning makes it hard to provide interpretability despite their improving prediction accuracy. That means that we are unable to explain how the models use features and samples to make decisions ( Molnar, 2020 ). The lack of interpretability has discouraged the application of machine learning methods to some extent. Therefore, many scholars in applied research still prefer to use simple models that are easy to interpret. To tackle this problem, some interpretation algorithms have been proposed, including linear regression and logit regression. For example, Local Interpretable Model-Agnostic Explanations (LIME) ( Ribeiro et al., 2016 ) and SHapley Additive exPlanations (SHAP) ( Lundberg et al., 2018 ) are usually used to interpret individual predictions; Partial Dependence Plot ( Friedman, 2001 ) and Accumulated Local Effects Plot ( Apley and Jingyu, 2016 ) are used to describe trends between variables and outcomes. Some studies used simple interpretable models (e.g., decision trees) to fit complex models and results in providing interpretation for the models ( Wan et al., 2020 ; Wu et al., 2020 ). This study used the SHAP method to provide an explanation of the trained model and carried out data-driven analysis to explore factors influencing reading literacy from a global perspective.

The Shapley value is a method from coalitional game theory that can be used in machine learning to measure the contribution of each variable when the model makes predictions for a particular instance, where contribution refers to the difference between the impact of a variable and the average impact. The Shapley value for each feature value is obtained by weighting and summing the marginal contributions of all possible combinations of feature values.

where S is a subset of the features used in the model and val ( S ) denotes the result of the prediction using the subset S .

When there are more features, the traversal of all combinations of features makes the computational process too complex. To address this problem, Lundberg et al. (2018) developed a fast and accurate algorithm for tree models, which reduces the computational complexity by computing all possible subsets in parallel while traversing all nodes along the decision path using the structural properties of the tree model. SHAP values, with their properties of local accuracy, missingness and consistency, can perform both local and global interpretability and are an effective method for explaining various machine learning models. SHAP values interpret the predicted values of the original model approximately as the sum of effects of all feature attributions.

where ϕ 0 is the predicted mean of all training samples. z ′ ∈ {0,1} p , where p is the number of features. z ′ i typically represents a feature being observed ( z ′ i = 1 ) or unknown ( z ′ i = 0 ), and ϕ i ’s are the effects of feature attributions. SHAP values are calculated separately for each variable of each sample, and the global contribution of each variable is obtained by taking the absolute average of the SHAP values of all samples of the same variable.

where N is the number of samples.

Descriptive statistics

The python 3.8 programs were used in the study. The results of descriptive statistics and correlation coefficient with PV1 ( M = 557.15, SD = 92.58) were shown in Table 1 . It was found that at the individual student level, the three reading metacognitions (understanding and remembering, summarizing, and assessing credibility) and reading interest were the most strongly correlated with PV1; at family level, ESCS and family wealth were the most correlated with PV1; features at school level were less correlated with PV1 compared to the other two, and it was total learning time, discipline climate in the subject, and teachers’ stimulation of students to engage in reading that show the highest correlations with PV1.

Table 1. Descriptive statistics for continuous variables.

Comparison of models

Three models were trained with 59 variables filtered to predict reading literacy PV1 values. To avoid overfitting, the hyperparameters of the three models mentioned above were optimized separately. A 10-fold cross-validation was performed in adjusting the parameters to select the optimal hyperparameter combinations on the test set. Table 2 shows the optimal hyperparameter combinations for the above three models, where the hyperparameter C of SVR is the relaxation variable penalty coefficient, gamma is inversely proportional to the standard deviation of the radial basis kernel function. RFR and XGBoost has more hyperparameters and using the grid search method would result in too much computation, so each hyperparameter has been optimized in turn in the order shown in Table 2 . The optimized predictive abilities of the three models are shown in Table 3 . The predictive ability were measured by the R 2 , Root Mean Square Error (RMSE) and Pearson’s correlation r. All coefficients were calculated after weighting and explained in detail as follows.

Table 2. Optimal hyperparameter combination of the three models.

Table 3. Comparison of the ability to predict and fit of the three models.

After 10-fold cross-validation, the prediction results of the three models were found to be similar, and XGBoost performed slightly better than the remaining two models. Therefore, the XGBoost model was chosen to fit all the data in this study, which was interpreted by the SHAP values.

Feature analysis

The SHAP values can calculate the contribution of each feature to each sample. Because features have both positive and negative effects on the sample, to measure the global importance of certain features, we take the absolute average of the SHAP values of the feature for all samples. The global contribution values of each variable are shown in Supplementary Appendix 1 . Figure 1 shows the top 20 features with the highest global contribution values. Consistent with the results of the descriptive statistics, the features at individual student level and at family level rank higher than those at school level, and the highest contributing feature at school level, “total learning times,” ranks only 10th. At individual student level, three reading metacognitive strategies were important predictors of reading literacy, in addition to reading interest, wellbeing, and expected occupational status; at family level, ESCS and the language used to communicate with family members were the most contributing features; and at school level, learning time and discipline in the classroom were the most important features. The SHAP summary plot was drawn to more clearly visualize the directionality of each feature’s influence on reading literacy, as shown in Figure 2 .

Figure 1. The average absolute SHAP value indicates the feature contribution.

Figure 2. SHAP summary plot.

Figure 2 shows the feature importance and direction of the effect of each feature on reading literacy. Each point in the figure is the SHAP value of a feature of an instance, with the position on the vertical axis indicating the feature value and the position on the horizontal axis indicating the SHAP value, and the feature values of all instances plotted with different color dots, where red dots and blue dots represents high feature values and low feature values, respectively. Taking the first feature, “metacognition: assess credibility” as an example, the SHAP value corresponding to the blue dot is negative, which means that the low feature value decreases reading literacy from the average of all samples, and the SHAP value corresponding to the red dot is positive, which means that the high feature value increases reading literacy, so the feature has a positive effect on reading literacy.

This study explored the most important predictors at the levels of the individual student, family, and school. First, at individual level, the strongest predictors of reading literacy were reading metacognition and reading interest. The global contributions of the three reading metacognitions were 20.127, 9.350, and 4.045, respectively, and the line chart of feature values and SHAP values were created to better visualize the effect of metacognitions on reading literacy (see Figure 3 ). As shown in Figure 3 , all three reading metacognitions showed a positive relationship with reading literacy, which is corresponding with the transition from blue to red in the Figure 2 . Therefore, reading metacognition had a positive predictive effect on reading literacy.

Figure 3. The line chart of meta-cognitions and Shap values.

Reading interest had a strong influence on students’ reading literacy, with a global contribution value of 9.176. As seen in Figure 2 , the blue points of reading interest (low reading interest) correspond to negative SHAP values, and the red points (high reading interest) to positive SHAP values. SHAP values increase with the growth of feature values. The scatter plot of interest and SHAP values (see Figure 4 ) also shows a positive relationship with reading literacy, with students who are more interested in reading being more likely to perform at a higher level of reading literacy.

Figure 4. The scatter plot of reading interest and Shap values.

The strongest predictors at family level are ESCS and home language environment. The global contribution value of ESCS is 12.708, the second strongest predictive feature after “metacognition: assess credibility.” As seen in Figure 2 , the color of this feature gradually transitions from blue to red along the positive horizontal axis, i.e., low ESCS corresponds to negative SHAP values, and high ESCS to positive ones. It can also be seen from the scatter plot of ESCS and SHAP values (see Figure 5 ) that there is a significant positive relationship between ESCS and reading literacy.

Figure 5. The scatter plot of ESCS and Shap values.

The categorical variable, which is one-hot coded in preprocessing, is the language used in communication with the mother and father, siblings, and classmates. The higher SHAP value is for the 0−1 variable generated from the fourth option in the questionnaire – “heritage language and test language are the same,” i.e., whether there is a dialect. The global contribution values for “heritage language and test language are the same” when communicating with mother, father, siblings, and classmates are 10.133, 6.124, 2.499, and 3.349, respectively. Because the homogeneity of variance was rejected by our data, Kruskal-Wallis test was used in the comparison among multiple groups. Table 4 shows the result of Kruskal-Wallis test for language used with mothers, fathers, siblings, and classmates, and reveals that the rank mean of category 4th is significantly higher than those of the other three categories. In addition, although the global contribution values for category 2 (“About equally often my heritage language and test language”) are small, at 0.094, 0.075, 0.069, and 0.013, the results of test shows that the mean rank for category 2th is significantly higher than the two remaining categories for language of communication with family members (“mostly my heritage language” and “mostly test language”). In summary, students without dialect have significantly higher reading literacy than others, while among the latter, students who use a dialect with the same frequency as Mandarin with family members have significantly higher reading literacy than those who use 0 only one of the languages.

Table 4. Result of Kruskal-Wallis test and rank mean for all groups for language used with family members and classmates.

School-level features generally lag behind individual level and home level features in the contribution ranking, with the highest contributions coming from learning time and discipline in the classroom. The global contribution values for total learning time and learning time (reading) are 5.306 and 2.646, respectively.

As seen in Figure 2 , the blue points for total learning hours correspond to negative SHAP values, indicating that less learning time reduces reading literacy, but the purple and red points overlap each other in the graph, indicating that the SHAP values for moderate learning time are approximately equal to those for high learning time. Drawing the scatter plot shown in Figure 6 , the relationship between total learning time and reading literacy is not simply positive and linear, indicating that SHAP values are less than zero, explaining why reading literacy of those sample are less than average, before 1,600 mins, then promote reading literacy from average after more than 1,600 mins. However, SHAP values first increase and then decrease until approximately 3,000 mins in which they are less than zero again. The results above indicates that prolonging time dedicated to learning to more than 1,600 mins makes no further contribution to improving reading literacy.

Figure 6. The scatter plot of total learning time and Shap values.

From the descriptive statistics, it was found that learning time (reading) shows a significant negative correlation with reading literacy at −0.075. From Figure 2 , it is known that SHAP values less than zero appear in red and blue at the same time, indicating that lower or higher learning time in reading reduces reading literacy. Plotting the scatter plot of this variable and the SHAP value in Figure 7 , its SHAP value has a clear peak at approximately 200–300 mins (approximately 6 h or so) per week when the SHAP value is greater than zero, but as reading time increases, the SHAP value begins to decline and keeps being greater than zero until 300 mins per week. Then the SHAP value less than zero has a negative effect on reading literacy. Therefore, there is an optimal length of total learning time and learning time in reading, which should be controlled within an appropriate range to avoid putting too much burden on students; otherwise, it would be detrimental to students’ academic performance.

Figure 7. The scatter plot of reading learning time and Shap values.

The global contribution of discipline climate is 2.064. As seen in Figure 2 , the blue points of discipline climate (more chaotic classroom climate) corresponds to negative SHAP values, and the red points (better classroom climate) to positive SHAP values. As seen from the scatter plot of discipline climate and it’s SHAP values in Figure 8 , discipline climate shows a positive relationship with reading literacy, indicating that it is more likely for students in orderly classroom to demonstrate higher reading literacy.

Figure 8. The scatter plot of discipline climate and Shap values.

Based on PISA 2018 dataset from four provinces/municipalities of China—Beijing, Shanghai, Jiangsu, and Zhejiang—this study used the XGBoost machine learning method to construct a prediction model and SHAP values to calculate the contribution of each feature to effectively predict reading literacy and interpret how it was influenced. Next, the influences that contribute the most to reading literacy were discussed at individual level, family level, and school level, respectively.

Individual level

Conclusion 1: Reading metacognition has a strong influence on reading literacy, with assessing credibility being the strongest predictor, summarizing the second, and understanding and remembering the weakest.

Based on the global contribution values calculated for each variable, reading metacognition was found to be an effective predictor, especially assessing credibility and summarizing. Reading metacognition refers to the extent to which the individual knows effective learning strategies ( Chunjin, 2020 ). Learning strategies can be divided into deep learning strategies, which refer to attempts to integrate new information with prior knowledge, and surface learning strategies, which involve repetitive rehearsal and rote memorization of information ( Murayama et al., 2013 ). Assessing credibility and summarizing can be categorized as deep learning strategies and understanding and remembering be regarded as surface learning strategies. The results indicates that the perception of effectiveness of deep learning strategies is far more important in the model than the perception of effectiveness of surface learning strategies because deep learning strategies promote connections between concepts and understanding of learning material; in contrast, surface learning strategies does not help students integrate new information into existing bodies of knowledge ( Murayama et al., 2013 ). Some studies ( Areepattamannil and Caleon, 2013 ; Murayama et al., 2013 ) suggested that remembering strategies shows an inverse relationship with mathematical literacy, but both Chunjin (2020) and the present study have found a positive predictive effect of remembering strategies on reading literacy, which may be due to the different disciplinary characteristics between reading and mathematics. Students’ ability to use strategies could regulate the level of language learning through cognitive strategies ( Ghafournia and Afghari, 2013 ), and the use of effective learning strategies also promotes students’ engagement in learning ( Lee and Shute, 2010 ); therefore, improving students’ ability to use cognitive strategies is crucial to improving reading proficiency. Paris and Oka (1986) argued that teachers can provide metacognitive knowledge about effective strategies to improve students’ reading skills, and Qishan et al. (2018) found that teacher’s instruction can influence students’ reading literacy not only directly but also indirectly through students’ learning strategies and reading metacognitive strategies. Therefore, teachers should realize that rote memorization does not improve students’ reading ability efficiently. What teachers should do is not only to lead students to read but also instruct them how to read, and the most effective way to improve students’ reading literacy is to help them master reading strategies and metacognitive strategies through teacher guidance.

Conclusion 2: Interest in reading can have a positive impact on reading literacy.

Reading interest is a significant positive predictor of reading literacy, which is consistent with the findings of many studies ( Wenjing and Tao, 2012 ; Lechner et al., 2019 ; Chunjin, 2020 ). In exploring how reading interest affects reading literacy, Schraw et al. (1995) argued that students with greater interest in a text always pay more attention to reading that text. Hongbo et al. (2016) argued that reading interest indirectly affects reading literacy by influencing reading engagement and that students who spend more time reading because of their interest are more likely to be good readers. If students are interested in reading, they are more likely to spend more attention, energy and time on learning and reading and, thus, gain more knowledge. It has also been found that the text itself also affects reading interest, and the relevance, vividness, and comprehensibility of the text are closely related to reading interest ( Schraw and Dennison, 1994 ). Schraw and Dennison (1994) found that the level of interest in a text is related to the student’s perspective and that a specific reading purpose increased the level of interest in that text; the more vivid and comprehensible the text is, the more interest of students will be aroused, and the higher the level of text recall, the more the knowledge gained. In addition, a number of studies have found that teacher support can influence reading engagement and reading interest and indirectly affect students’ reading literacy ( Kaili and Wei, 2021 ). Therefore, teachers should provide appropriate support from various aspects, such as emotional care, learning instruction, ability guidance, in addition to focusing on cultivating students’ reading interests in reading teaching practice, taking the vividness of reading materials into consideration when selecting reading texts, giving instruction in conjunction with students’ interests, and helping students understand the relevance of textual knowledge in an interest-oriented manner to increase their reading level.

Family level

Conclusion 1 : High family ESCS is advantageous for reading literacy.

This study found that the ESCS is a strong predictor on reading literacy, and the higher the ESCS, the more beneficial it was for reading literacy. Welch (2013) , Silin et al. (2014) also pointed out that there is a significant relationship between family financial level and academic performance, with children from families with high ESCS having an advantage over those with low ESCS. Tatlisu et al. (2011) , Banerjee and Lamb (2016) even argued that family background has a much greater impact on students than school influence, further illustrating the importance of family factors. Low-ESCS families usually have less time, resources, or means to engage in their children’s education, and parents place less emphasis on books, reading, and education ( Banerjee and Lamb, 2016 ). In contrast, high-ESCS families usually possess more cultural capital and a greater ability to create a good learning environment ( Silin et al., 2014 ), so that their children usually have more opportunities to devote themselves to reading and learning. Even when their children do not do well in school, high-ESCS families are more able to afford high tuition and provide extracurricular learning resources to assist their children’s learning ( Mauldin et al., 2001 ). Families with high ESCS have more educational advantages than families with low ESCS, and over time, advantaged families would occupy more quality educational resources and opportunities, and disadvantaged families would be increasingly disadvantaged ( Silin et al., 2014 ), which goes against the principle of educational equity. Although studies have shown that school interventions cannot completely eliminate the gap caused by family financial situations ( Goldfeld et al., 2021 ), student reading strategies, interest in reading, school operations, and teacher-student relationships have been proved to have a significant contribution to the academic performance of low ESCS students ( Hao and Yifang, 2020 ; Mingman and Guomin, 2021 ).

Conclusion 2: Home language environment has an impact on students’ reading literacy, with dialects having a negative effect on reading literacy; however, proficiency in both a dialect and Mandarin has a positive effect on reading literacy.

The results indicate that home language environment has a strong influence on students’ reading literacy. After one-hot coding of the language variables used, it is found that when the “heritage language and test language are the same” (i.e., whether there is a dialect), there was a greater impact of the home language environment on reading literacy. This study found that dialects negatively influence students’ reading literacy. Due to the difference between spoken or written dialects and the official lingua franca, students whose native language is a dialect often make grammatical mistakes and showed poor expressions when learning Mandarin, which can also hinder their written reading and writing ( Qin and Zhe, 2020 ; Yinyin, 2020 ). Some studies have shown that children with a Mandarin dialect perform worse in the concepts of word (awareness of words as separate units) and syntactic awareness (syntactic knowledge and competence) than Mandarin monolingual children ( Hanling and Rongbao, 2014 ; Yongxiang and Rongbao, 2015 ). However, dialects are not always harmful, and Hanling and Rongbao (2014) noted that dialect experience has a positive effect on the development of children’s sense of denotational arbitrariness (the extent to which children understand the conventional relationship between language as a symbol and the meaning it refers to) due to bilingual children’s greater metalinguistic awareness, i.e., their ability to distinguish between two language systems early in language learning and to suppress non-target language during language use ( Wenyu, 2010 ), which allows them to perform better than monolingual children on all metalinguistic tasks requiring a high degree of controlled processing and to generalize the advantage of controlled attention to other nonlinguistic domains ( Bialystok, 1988 ). In addition, bilingual children had different language learning strategies and could learn new languages faster than monolingual learners ( Nayak et al., 1990 ). Bialystok (1988) suggested that these advantages of bilingual children are contingent on a good understanding of the second language and becoming “balanced” bilinguals, which explained why the negative effect of dialect often plays out in the early years of language learning, and the disadvantage of dialect-common speaking children disappear when children moved into higher grades ( Sumei et al., 2013 ). This was also supported by our study, which found that students in all three categories, except “heritage language and test language are the same,” using both dialect and Mandarin with family members rather than classmates have significantly higher reading performance than students who use only one language because the former are closer to the “balanced” bilinguals. In China, students usually are required to speak Mandarin in school, so we suggest parents, in dialect areas, to not ignore teaching of dialects and provide children with opportunities to practice dialect at home so that kids could become “balanced” bilinguals and give full play to the lingual advantages of bilingual children.

School level

Conclusion 1: Learning time positively predicted reading literacy, but longer learning time does not improve academic performance and may even be the reverse.

Some studies ( Yan and Leung, 2012 ; Kidron and Lindsay, 2014 ; Jez and Wassmer, 2015 ) indicated that longer learning times improves students’ academic performance, and the present study gets a similar conclusion. However, the difference is that the present study found that increasing learning time is of limited help to improving achievement, with optimal total learning time for improving academic performance being about 1,600 mins or 26.6 h per week and learning time in reading 200 to 300 mins or 6 h per week. Exceeding this time limits would may be detrimental to academic performance instead of improving it. This is consistent with the findings of Xuejun (2014) , Fei and Hongbo (2017) that, although learning time is positively correlated with academic performance, too much time spent on learning is not necessarily better. Bloom (1974) distinguished between “elapsed time” (time presumably working on the task) and the actual time spent on the task, with the latter closely related to learning achievement. Extending learning time does not necessarily increase the time spent on work but rather increased students’ coursework burden ( Linchun and Lujian, 2007 ; Mengjie and Tao, 2019 ), which leads to negative emotions such as anxiety and aversion to learning ( Xiao, 2013 ), which may lead students to flee and avoid learning, resulting in negative academic performance ( Arsenio and Loria, 2014 ). Our study suggests that teachers and parents should realize that reducing students’ play and rest time to unnecessarily extend learning time does not do much for improving academic performance. Chinese officials have begun to call for reducing the burden of homework and out-of-school training for students in compulsory education, and teachers should reasonably control class time and learning time to effectively help students improve their academic performance.

Conclusion 2: Maintaining discipline climate was useful for improving reading literacy.

This study has found that discipline climate was a significant positive predictor of reading literacy, which was in line with the findings of Arens et al. (2015) , Simba et al. (2016) and others. Discipline played a very important role in students’ academic performance; students with a sense of discipline are more focused, study harder, show greater determination and are more likely to be accepted and appreciated by teachers and parents, thus develop more positive self-identity and have more motivation ( Simba et al., 2016 ), while disruptions in classroom order could be detrimental to teaching efficiency and may even interrupt normal teaching and learning activities. Some studies have shown that teachers’ content, delivery process and teaching attitude are important factors in students’ classroom disruptions ( Guiping et al., 2005 ), and that most classroom disrupters are students who are not engaged in learning, are bored and, as a result, would miss learning opportunities; furthermore, their dissatisfaction with learning exacerbated disengagement ( Rahimi and Karkami, 2015 ; Lopes and Oliveira, 2017 ). Therefore, in addition to acquiring necessary academic and pedagogical skills, teachers need to acquire management skills to ensure order in the class ( Lopes and Oliveira, 2017 ). While legitimate use of punishment helps to maintain the authority of discipline and could have the effect of maintaining discipline in the short term, overreliance on punishment strategies cannot necessarily ensure long-term stability in class order ( Mark, 2006 ) and may even exacerbate classroom disruptions and stimulate student resistance and hostility ( Roache and Lewis, 2011 ). Sun (2015) held that setting transparent rules, talking after class, using punishment strategies appropriately, building positive relationships with students who were mutually trusted, and promoting student engagement in learning were effective discipline strategies. Maintaining order in class was a major concern for most new teachers ( Sutton and Wheatley, 2003 ); thus, not only instruction in teaching competencies but also effective discipline strategies should be included in preservice training for teachers to maintain order in the classroom and ensure that instructional activities could be carried out properly.

Conclusion and limitations

Based on data from four provinces/municipalities of China in PISA 2018, this study explored the most effective predictors for reading literacy using the XGBoost model and the SHAP values and found that individual-level and family level features had a more significant impact on reading literacy. Reading metacognition at the individual level was the strongest predictor of all variables, in addition to reading interest as an effective positive influence, indicating that guiding students to master appropriate reading strategies and stimulating reading interest are useful for improving students’ reading literacy. Family level ESCS and family language environment are effective predictors of reading literacy: ESCS has a strong influence and positively predicted reading literacy; speaking a dialect was detrimental for reading literacy, although “balanced bilinguals” who were proficient in both a dialect and Mandarin had an advantage over monolingual students. At school level, there was an optimal value for total learning time and reading learning time. Extending learning time cannot improve academic performance, may play a negative role instead; thus, parents and teachers should reasonably control class time and learning time. In addition, discipline climate is beneficial for students’ reading literacy.

There were several limitations of this study that need to be noted. First, because the data in this study are cross-sectional, it is difficult to determine whether there is a causal relationship between the variables, and the variable contributions derived from this study are only predictive contributions rather than causal analyses. The variable attributions through SHAP values make it possible to simply examine a variable in isolation, but in fact, these variables have complex interactions and causal networks. For instance, school and family factors not only have an impact on reading literacy, but also affect individual-level factors. Therefore, it is recommended that future research focus on the mechanisms of influence between variables. Second, while this study obtained some valuable and interesting conclusions using XGBoost, there may be a risk of overfitting using machine learning as the hierarchical data contains multiple levels. Therefore, considering the hierarchical characteristics of PISA data, Future analysis may consider combining machine learning with a mixed effects model to further validate the conclusions of our study. Third, the data used in this study was made up of four provinces/municipalities of China, which were the most developed regions in China and was not representative of students across the country. Forth, considering that among the four provinces/municipalities of China, Shanghai, Jiangsu, and Zhejiang were dialect regions, the high effect of language variables in this study may actually be attributed to regional differences.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: https://www.oecd.org/pisa/data/2018database/ .

Author contributions

HL provided research guidance and wrote the manuscript. XC analyzed the data and wrote the data analysis reports. XL polished and revised the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported in part by the National Social Science Fund of China under Grant 20CTJ019.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyg.2022.948612/full#supplementary-material

Ahmed, W., Minnaert, A., Werf, G., and Kuyper, H. (2010). Perceived social support and early adolescents’ achievement: The mediational roles of motivational beliefs and emotions. J. Youth Adolesc. 39, 36–46. doi: 10.1007/s10964-008-9367-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Apley, D. W., and Jingyu, Z. (2016). Visualizing the effects of predictor variables in black box supervised learning models. arXiv [Preprint]

Google Scholar

Areepattamannil, S., and Caleon, I. S. (2013). Relationships of cognitive and metacognitive learning strategies to mathematics achievement in four high-performing east asian education systems. J. Genet. Psychol. 174, 696–702. doi: 10.1080/00221325.2013.799057

Arens, A. K., Morin, A. J. S., and Watermann, R. (2015). Relations between classroom disciplinary problems and student motivation: Achievement as a potential mediator? Learn. Instr. 39, 184–193. doi: 10.1016/j.learninstruc.2015.07.001

CrossRef Full Text | Google Scholar

Arsenio, W. F., and Loria, S. (2014). Coping with negative emotions: Connections with adolescents’ academic performance and stress. J. Genet. Psychol. 175, 76–90. doi: 10.1080/00221325.2013.806293

Banerjee, P. A., and Lamb, S. (2016). A systematic review of factors linked to poor academic performance of disadvantaged students in science and maths in schools. Cogent Educ. 3, 1–17. doi: 10.1080/2331186X.2016.1178441

Berkowitz, R. (2021). School climate and the socioeconomic literacy achievement gap: Multilevel analysis of compensation, mediation, and moderation models. Child. Youth Serv. Rev. 130, 1–10. doi: 10.1016/j.childyouth.2021.106238

Bialystok, E. (1988). Levels of bilingualism and levels of linguistic awareness. Dev. Psychol. 24, 560–567. doi: 10.1037/0012-1649.24.4.560

Bloom, B. S. (1974). Time and learning. Am. Psychol. 29, 682–688. doi: 10.1037/h0037632

Bo, Y., Decheng, Z., and Luhuan, W. (2017). What factors are influencing the academic performance of Chinese students?- -Multi-level analysis based on PISA2015 data of students in four provinces (municipalities) in China. Sch. Admin. 9–12.

Bosch, N. (2021). Identifying supportive student factors for mindset interventions: A two-model machine learning approach. Comput. Educ. 167, 1–15. doi: 10.1016/j.compedu.2021.104190

Buckingham, J., Wheldall, K., and Beaman-Wheldall, R. (2013). Why poor children are more likely to become poor readers: The school years. Aust. J. Educ. 57, 190–213. doi: 10.1177/0004944113495500

Chen, T., and Guestrin, C. (2016). “XGBoost: A scalable tree boosting system,” in Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining , (New York, NY: ACM). doi: 10.1145/2939672.2939785

Chunjin, C. (2020). New findings in reading literacy assessment among students in the four provinces/municipalities of China in PISA 2018. J. East China Norm. Univ. Educ. Sci. 38, 22–61.

Crosnoe, R., Leventhal, T., Wirth, R. J., Pierce, K. M., Pianta, R. C., and Network, N. E. C. C. (2010). Family socioeconomic status and consistent environmental stimulation in early childhood. Child Dev. 81, 972–987. doi: 10.1111/j.1467-8624.2010.01446.x

Davis-Kean, and Pamela, E. (2005). The influence of parent education and family income on child achievement: The indirect role of parental expectations and the home environment. J. Fam. Psychol. 19, 294–304. doi: 10.1037/0893-3200.19.2.294

Deary, I. J., Strand, S., Smith, P., and Fernandes, C. (2007). Intelligence and educational achievement. Intelligence 35, 13–21. doi: 10.1016/j.intell.2006.02.001

Dreher, J. M. M. S., and Mikulecky. (2000). What will be the demands of literacy in the workplace in the next millennium? Read. Res. Q. 35, 378–383. doi: 10.1598/RRQ.35.3.3

Fei, H., and Hongbo, W. (2017). Research on secondary school students’ optimal out-of-class study time: Based on PISA2012 data. Prim. Second. Sch. Abroad 23–30.

Fengning, S., Ge, S., Xianjun, S., and Biyin, Z. (2000). Relationship between middle school students’ reading motivation and the amount and achievement of their reading. J. Psychol. Sci. 23, 84–87.

Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Ann. Stat. 29, 1189–1232. doi: 10.1214/aos/1013203451

Ghafournia, N., and Afghari, A. (2013). The interaction between reading comprehension cognitive test-taking strategies, test performance, and cognitive language learning strategies. Procedia Soc. Behav. Sci. 70, 80–84. (Complete). doi: 10.1016/j.sbspro.2013.01.041

Goldfeld, S., Moreno-Betancur, M., Guo, S., Mensah, F., and O’Connor, M. (2021). Inequities in children’s reading skills: The role of home reading and preschool attendance. Acad. Pediatr. 21, 1046–1054. doi: 10.1016/j.acap.2021.04.019

Guiping, W., Xiaoyan, S., Ruifang, G., and Yan, L. (2005). Review on the foreign studies on classroom discipline. Stud. Foreign Educ. 32, 77–80.

PubMed Abstract | Google Scholar

Guryan, J., Hurst, E., and Kearney, M. (2008). Parental education and parental time with children. J. Econ. Perspect. 22, 23–46. doi: 10.1257/jep.22.3.23

Halle, T. G., Kurtz-Costes, B., and Mahoney, J. L. (1997). Family influences on school achievement in low-income, African American children. J. Educ. Psychol. 89, 527–537. doi: 10.1037/0022-0663.89.3.527

Han, L. (2017). The effects of family income on children’s education: An empirical analysis of CHNS data. Res. Mod. High. Educ. 4, 49–54.

Hanling, W., and Rongbao, L. (2014). The effect of dialect experience on children’s development of word awareness. Mod. Foreign Lang. (Bimonthly) 37, 805–814.

Hao, L., and Yifang, Z. (2020). The influencing factors of disadvantaged students’ reading literacy: The HLM analysis basaed on the China’s PISA 2018 data. China Examinations 42–48.

Hart, S. A., Soden, B., Johnson, W., Schatschneider, C., and Taylor, J. (2013). Expanding the environment: Gene×school-level SES interaction on reading comprehension. J. Child Psychol. Psychiatry 54, 1047–1055. doi: 10.1111/jcpp.12083

Hongbo, W., Kaili, L., and Xianwei, L. (2016). Effect of family environment on reading ability: The mediating effects of reading engagement and reading interest among junior high school students. Acta Psychol. Sin. 248–257. doi: 10.3724/SP.J.1041.2016.00248

Hsu, C. K., Hwang, G. J., and Chang, C. K. (2013). A personalized recommendation-based mobile learning approach to improving the reading performance of EFL students. Comput. Educ. 63, 327–336. doi: 10.1016/j.compedu.2012.12.004

Hu, J., Dong, X., and Peng, Y. (2022). Discovery of the key contextual factors relevant to the reading performance of elementary school students from 61 countries/regions: Insight from a machine learning-based approach. Read. Writ. 35, 93–127. doi: 10.1007/s11145-021-10176-z

Janet, C., and Enrico, M. (2003). Mother’s education and the intergenerational transmission of human capital: Evidence from college openings and longitudinal data. Q. J. Econ. 118, 1495–1532. doi: 10.1162/003355303322552856

Jez, S. J., and Wassmer, R. W. (2015). The impact of learning time on academic achievement. Educ. Urban Soc. 47, 284–306. doi: 10.1177/0013124513495275

Jing, L. (2012). The Influence of engagement in reading and learning strategy on reading performance: Evidence-based research with shanghai PISA 2009 data. Res. Educ. Dev. 17–24.

Kaili, L., and Wei, T. (2021). The effect of teacher support on reading literacy of middle school students:The mediating role of reading interest and reading engagement. Chin. J. Spec. Educ. 90–96.

Kidron, Y., and Lindsay, J. (2014). The effects of increased learning time on student academic and nonacademic outcomes: Findings from a metaanalytic review. Washington, DC: U.S. Department of Education.

Law, Y. K. (2011). The role of teachers’ cognitive support in motivating young Hong Kong Chinese children to read and enhancing reading comprehension. Teach. Teach. Educ. 27, 73–84. doi: 10.1016/j.tate.2010.07.004

Lechner, C. M., Ai, M., and Knopf, T. (2019). Should students be smart, curious, or both? Fluid intelligence, openness, and interest co-shape the acquisition of reading and math competence. Intelligence 76, 1–12. doi: 10.1016/j.intell.2019.101378

Lee, J., and Shute, V. J. (2010). Personal and social-contextual factors in K–12 academic performance: An integrative perspective on student learning. Educ. Psychol. 45, 185–202. doi: 10.1080/00461520.2010.493471

Linchun, T., and Lujian, F. (2007). Empirical research string on the relationship between academic burden and academic performance. J. Shanghai Educ. Res. 32–36.

Logan, S. (2008). An examination of the effects of reading instruction and gender differences on children’s reading. Doctor of Philosophy. Kingston upon Hull: The University of Hull.

Logan, S., Medford, E., and Hughes, N. (2011). The importance of intrinsic motivation for high and low ability readers’ reading comprehension performance. Learn. Individ. Differ. 21, 124–128. doi: 10.1016/j.lindif.2010.09.011

Lopes, J., and Oliveira, C. (2017). Classroom discipline: Theory and practice classrooms: Academic content and behavior strategy instruction for students with and without disabilities. Nova Sci. 2, 231–253. (Reprinted.

Lundberg, S. M., Erion, G. G., and Lee, S. I. (2018). Consistent individualized feature attribution for tree ensembles. arXiv [Preprint]

Luo, S., Song, N., and Wang, Y. (2016). An evidence-based study of reading literacy in elementary school: Connotations, values and expressions. J. Chin. Soc. Educ. 77–83.

Mark, C. (2006). Managing school discipline and implications for school social workers: A review of the literature. Child. Sch. 28, 219–227. doi: 10.1093/cs/28.4.219

Mauldin, T., Mimura, Y., and Lino, M. (2001). Parental expenditures on children’s education. J. Fam. Econ. Issues 22, 221–241. doi: 10.1023/A:1016647806016

Mengjie, H., and Tao, Y. (2019). On classification model of compulsory education student academic burden and its application. Educ. Sci. 35, 1–6.

Mingman, R., and Guomin, Z. (2021). Empirical study on the influencing factors of adverse reading resistance for disadvantaged students. Educ. Sci. Res. 72–76.

Molnar, C. (2020). Interpretable machine learning. Morrisville, NC: Lulu Press INC.

Ms, A., and Syh, B. (2021). Teacher support and academic engagement among EFL learners: The role of positive academic emotions. Stud. Educ. Eval. 70, 1–8. doi: 10.1016/j.stueduc.2021.101060

Muntoni, F., and Retelsdorf, J. (2018). Gender-specific teacher expectations in reading—The role of teachers’ gender stereotypes. Contemp. Educ. Psychol. 54, 212–220. doi: 10.1016/j.cedpsych.2018.06.012

Murayama, K., Pekrun, R., Lichtenfeld, S., and Hofe, R. V. (2013). Predicting long-term growth in students’ mathematics achievement: The unique contributions of motivation and cognitive strategies. Child Dev. 84, 1475–1490. doi: 10.1111/cdev.12036

Nayak, N., Hansen, N., Krueger, N., and Mclaughlin, B. (1990). Language-learning strategies in monolingual and multilingual adults. Lang. Learn. 40, 221–244. doi: 10.1111/j.1467-1770.1990.tb01334.x

Ninaus, M., Greipl, S., Kiili, K., Lindstedt, A., Huber, S., Klein, E., et al. (2019). Increased emotional engagement in game-based learning – A machine learning approach on facial emotion detection data. Comput. Educ. 142, 1–10. doi: 10.1016/j.compedu.2019.103641

O’Malley, M., Voight, A., Renshaw, T. L., and Eklund, K. (2015). School climate, family structure, and academic achievement: A study of moderation effects. Sch. Psychol. Q. 30, 142–157. doi: 10.1037/spq0000076

OECD (2009a). “Analyses with plausible values,” in PISA data analysis manual: SPSS, (second edition) , (Paris: OECD Publishing), doi: 10.1787/9789264056275-9-en

OECD (2009b). “Analyses with plausible values,” in PISA data analysis manual: SPSS, (second edition) , (Paris: OECD Publishing), doi: 10.1787/9789264056275-en

OECD (2010). PISA 2009 assessment framework. Paris: OECD Publishing, doi: 10.1787/9789264062658-en

OECD (2013). OECD skills outlook 2013. Paris: OECD Publishing, doi: 10.1787/9789264204256-en

OECD (2019). PISA 2018 assessment and analytical framework. Paris: OECD Publishing, doi: 10.1787/b25efab8-en

Paris, S., and Oka, E. (1986). Children’s reading strategies, metacognition, and motivation. Dev. Rev. 6, 25–56. doi: 10.1016/0273-2297(86)90002-X

Perparim, S. (2014). School effects, gender and socioeconomic differences in reading performance: A multilevel analysis. Int. Educ. Stud. 7, 28–39. doi: 10.5539/ies.v7n11p28

Pingping, Z., Lingmei, L., and Tao, X. (2011). Cross-cultural comparison of linkage between school climate and students’ mathematical achievement: A multilevel analysis. Psychol. Dev. Educ. 27, 625–632.

Pokay, P., and Blumenfeld, P. C. (1990). Predicting achievement early and late in the semester: The role of motivation and use of learning strategies. J. Educ. Psychol. 82, 41–50. doi: 10.1037/0022-0663.82.1.41

Qin, L., and Zhe, W. (2020). Factors influencing students’ reading literacy in China–analysis based on PISA2018 data. Shanghai Res. Educ. 24–29.

Qishan, C., Yaying, L., Zhonglin, W., Shutong, L., Jieying, L., and Yurou, K. (2018). Teacher support, reading strategy and reading literacy: A two-level mediation model. Glob. Educ. 47, 51–61.

Rahimi, M., and Karkami, F. H. (2015). The role of teachers’ classroom discipline in their teaching effectiveness and students’ language learning motivation and achievement: A path method. Iran. J. Lang. Teach. Res. 3, 57–82.

Retelsdorf, J., Ller, K. O., and Ller, M. J. (2011). On the effects of motivation on reading performance growth in secondary school. Learn. Instr. 21, 550–559. doi: 10.1016/j.learninstruc.2010.11.001

Retelsdorf, J., Schwartz, K., and Asbrock, F. (2015). Michael can’t read!” Teachers’ gender stereotypes and boys’ reading self-concept. J. Educ. Psychol. 107, 186–194. doi: 10.1037/a0037107

Ribeiro, M. T., Singh, S., and Guestrin, C. (2016). “Why should i trust you?”: Explaining the predictions of any classifier,” in Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining , (New York, NY: ACM). doi: 10.1145/2939672.2939778

Roache, J., and Lewis, R. (2011). Teachers’ views on the impact of classroom management on student responsibility. Aust. J. Educ. 55, 132–146. doi: 10.1177/000494411105500204

Schraw, G., and Dennison, R. S. (1994). The effect of reader purpose on interest and recall. J. Read. Behav. 26, 1–18. doi: 10.1080/10862969409547834

Schraw, G., Bruning, R., and Svoboda, C. (1995). Sources of Situational Interest. J. Lit. Res. 27, 1–17. doi: 10.1080/10862969509547866

Sheng, Z., Mei, S., Lili, W., and Hongbo, W. (2014). The effect of teachers on students’ reading ability: The mediating effect of reading engagement. Chin. J. Spec. Educ. 684–689.

Silin, H., Ziqiang, X., and Jiawei, H. (2014). The relationship between family background and higher education enrollment opportunities and its trends. China Youth Study 46–52.

Simba, N. O., Agak, J. O., and Kabuka, E. K. (2016). Impact of discipline on academic performance of pupils in public primary schools in Muhoroni sub-county, Kenya. J. Educ. Pract. 7, 164–173.

Smith, J. K., Smith, L. F., Gilmore, A., and Jameson, M. (2012). Students’ self-perception of reading ability, enjoyment of reading and reading achievement. Learn. Individ. Differ. 22, 202–206. doi: 10.1016/j.lindif.2011.04.010

Spagat, M. (2006). Human capital and the future of transition economies. J. Comp. Econ. 34, 44–56. doi: 10.1016/j.jce.2005.11.002

Sumei, C., Rongbao, L., Guangze, L., Youkun, W., and Liqiong, W. (2013). The effect of dialect experience on Chinese children’s Mandarin phonological awareness. Read. Writ. 26, 1317–1335. doi: 10.1007/s11145-012-9420-5

Sun, R. (2015). Teachers’ experiences of effective strategies for managing classroom misbehavior in Hong Kong. Teach. Teach. Educ. 46, 94–103. doi: 10.1016/j.tate.2014.11.005

Sutton, R. E., and Wheatley, K. F. (2003). Teachers’ emotions and teaching: A review of the literature and directions for future research. Educ. Psychol. Rev. 15, 327–358. doi: 10.1023/A:1026131715856

Swalander, L., and Taube, K. (2007). Influences of family based prerequisites, reading attitude, and self-regulation on reading ability. Contemp. Educ. Psychol. 32, 206–230. doi: 10.1016/j.cedpsych.2006.01.002

Tatlisu, M. A., Ozcan, K. S., Güngör, B., Ekmekçi, A., and Erdinler, I. (2011). Family background and school effects on student achievement: A multilevel analysis of the coleman data. Coron. Artery Dis. 25, 97–132. doi: 10.1177/016146811111300101

Ülle, S., Kari, N., and Antero, M. (2015). Learning strategies explaining differences in reading proficiency. Findings of Nordicand Baltic countries in PISA 2009. Learn. Individ. Differ. 42, 36–43. doi: 10.1016/j.lindif.2015.08.025

Unrau, N., and Schlackman, J. (2006). Motivation and its relationship with reading achievement in an urban middle school. J. Educ. Res. 100, 81–101. doi: 10.3200/JOER.100.2.81-101

Wan, A., Dunlap, L., Ho, D., Yin, J., Lee, S., Jin, H., et al. (2020). “NBDT: Neural-backed decision trees,” in Proceedings of the international conference on learning representations (ICLR 2021) , Vienna.

Webster, B. J., and Fisher, D. L. (2003). School-level environment and student outcomes in mathematics. Learn. Environ. Res. 6, 309–326. doi: 10.1023/A:1027383925394

Welch, A. L. (2013). The effect of poverty on the achievement of urban african american male students successfully completing high school. Doctor of Education. Omaha: The Graduate College of the University of Nebraska.

Wenjing, Z., and Tao, X. (2012). A cross-culture comparison study of the impact of reading engagement on reading literacy: Based on PISA 2009. Psychol. Dev. Educ. 28, 175–183.

Wenyu, W. (2010). Language ability development of bilingual children a review of empirical studies. Foreign Lang. Res. 52–60.

Wu, M., Parbhoo, S., Hughes, M., Kindle, R., Celi, L., Zazzi, M., et al. (2020). Regional tree regularization for interpretability in deep neural networks. Proc. AAAI Conf. Artif. Intell. 34, 6413–6421. doi: 10.1609/aaai.v34i04.6112

Xiao, J. (2013). Academic stress, test anxiety, and performance in a chinese high school sample: The moderating effects of coping strategies and perceived social support. Doctor of Philosophy. Atlanta, GA: College of Education Georgia State University.

Xiaoying, X. (2022). Family income, parental education and chinese preschoolers’ cognitive school readiness: Authoritative parenting and parental involvement as chain mediators. Front. Psychol. 13, 1–12. doi: 10.3389/fpsyg.2022.745093

Xing, W. D. (2019). Dropout prediction in MOOCs: Using deep learning for personalized intervention. J. Educ. Comput. Res. 57, 547–570. doi: 10.1177/0735633118757015

Xuejun, S. (2014). Research on Shanghai secondary school students’ optimal learning time based on pisa data. Res. Educ. Dev. 9–14.

Yan, Z., and Leung, F. (2012). Homework and mathematics achievement in Hong Kong: Evidence from the timss 2003. Int. J. Sci. Math. Educ. 10, 907–925. doi: 10.1007/s10763-011-9302-3

Yinyin, W. (2020). Research on rural children’s reading ability of multiple factors interactions influence by data mining approach. Inf. Res. 74–81.

Yongxiang, H., and Rongbao, L. (2015). The effect of dialect experience on the development of children’s syntactic awareness. J. Foreign Lang. 38, 58–67.

Keywords : reading literacy, Influencing factors, XGBoost, SHAP, interpretability\keywordbelowspace-30pt

Citation: Liu H, Chen X and Liu X (2022) Factors influencing secondary school students’ reading literacy: An analysis based on XGBoost and SHAP methods. Front. Psychol. 13:948612. doi: 10.3389/fpsyg.2022.948612

Received: 20 May 2022; Accepted: 29 August 2022; Published: 23 September 2022.

Reviewed by:

Copyright © 2022 Liu, Chen and Liu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Hao Liu, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Factors Affecting Reading Comprehension

This article explores factors affecting reading comprehension. It discusses different concepts on reading comprehension and describes some components of reading abilities, which are important to decode the written text. Large amount of research was devoted to analyze linguistic, cognitive and other factors influencing reading abilities. This study highlights the role of vocabulary, phonological, morphological awareness and metalinguistic knowledge in reading comprehension.

Related Papers

Dr. Khalid Albadawi

The aim of this study was to investigate the influence of reading comprehension in reading fluency. The lack of teaching reading comprehension effect on learning reading fluency. This study aims at determining the relationship between reading comprehension and reading fluency. Two subject groups represent different situations were selected for an investigation. An instrument for data collection and statistical analysis to gain results. These instruments are pre and posttests. Certain procedures were followed in conducting the pre and posttests. The two scales of reliability and validity of the tests were used to achieve consistent answers. The above instruments helped draw significant findings. These findings proved the importance of reading comprehension in reading fluency and explained the relationship among the English reading fluency and the effective ways of teaching English reading comprehension. In addition, the study emphasized that most of reading fluency difficulties can b...

Peter GAVORA

Educational Sciences Theory&Practice

This study examined the relationship between fourth grade primary school students’ reading habits/conditions/ situations and their comprehension regarding what they read. For this purpose, a correlational survey method was used in the study. 90 fourth-grade students who were attending a state primary school in the center of Kütahya participated to the study. Firstly, there are four separate tests: a fill in the blank test which measures reading as a process, a short answer test which measures remembering, multiple choice test which measures both superficial and in-depth meaning linking and open-ended questions which measure meaning linking skills ) were administered to the students. Then, students were asked to read loudly 409-word narrative text and students’ voices were recorded. The records were analyzed by experts to determine the number of words that students read per minute, students’ reading mistakes and their prosody. Pearson correlation analysis was used to determine the relationship between fluent reading skills and reading comprehension. Moreover, multiple regression analysis was conducted to examine the predictive power of reading skills for comprehension. The findings of the study demonstrated that fluent reading was an indicator of comprehending; prosody predicted in-depth meaning linking better than the other fluent reading skills; correct reading skills predicted superficial meaning linking better. The findings also revealed that there was a weak correlation between reading speed and comprehension. Educational Sciences: Theory & Practice - 13(4) • 2287-2290 ©2013 Educational Consultancy and Research Center

International Journal of Innovative Language, Literature & Art Studies 11(4):48-54, Oct.-Dec., 2023

SEAHI Global Publications SEAHI

The aim of reading comprehension is to derive meaning from a given text, while reading comprehension is decoding and constructing meaning through interaction with a given text. This paper reviews the concept of reading comprehension, levels of text comprehension in oral reading, the types of oral reading comprehension. The paper further discussed reading comprehension difficulties, the causes of reading comprehension difficulties, the strategies for teaching reading comprehension as well as the strategies for assessing oral reading comprehension. The paper recommends that teachers should teach skills that lead to text comprehension more specifically from the early classes.

Fatihcetin Cetinkaya

Educational Sciences Theory and Practice

Mustafa Kaan Basaran

Shanu Singh

United International Journal for Research & Technology (UIJRT)

UIJRT | United International Journal for Research & Technology

The study determined the factors that affect the reading comprehension of Grade 7 students in Magallanes National High School, Magallanes, Sorsogon for school year 2019-2020. A mixed method research approach was used to understand and determine the factors that affect the reading comprehension of Grade 7 students. The participants of the study were the 148 Grade 7 students of Magallanes National High School. In order to determine the reading comprehension level of students in the pre-test and post-test, pre-test and post-test was used. To determine the difference between the results of pre-test and post-test, T-test of paired samples was used. Likewise, to gather relevant information on the factors that may affect the reading comprehension of students an interview schedule was used. Moreover, the data gathered were analyzed using frequency count, mean scores, and Descriptive Phenomenology method. The study showed that there are 14 non-readers, 134 students under frustration level, 136 under instructional level and 6 under independent level in the pre-test. There are 14 non-readers, 14 students under frustration level, 171 under instructional level and 91 students under independent level in the post-test. Likewise, one of the factors that affects the reading comprehension of students is the lack of reading materials was one of the factors affecting the reading comprehension of students, wherein students are provided with limited materials for reading. Another factor is that teachers do not conduct monitoring and home visitation. Students were not monitored or visited especially those who need reading remediation. Moreover, the reading difficulties experienced by students such as understanding the text in English, sometimes difficulty understanding the reading text that is long and words that are longer and having a few knowledge about the reading text are also considered factors. Furthermore, the computed t value on the difference between the results of pre-test and post-test is 5.05. Moreover, to improve the reading comprehension of students, a workbook entitled “ON YOUR MARKS, GET SET, READ: A Workbook for Grade 7 Students” was proposed. In the light of the findings, the present study recommended that teachers may use appropriate learning materials and strategies and may provide reading activities that are relevant for the students to enrich and enhance their reading skills particularly in reading comprehension. Likewise, teachers should conduct monitoring and home visitation to struggling readers and they may collaborate with each other to provide instructional materials (print or non-print) and conduct relevant reading activities for the students. To improve the reading comprehension of students, teachers may employ strategies and may utilize appropriate reading materials. that may improve Another, there is a need to utilize and implement the developed reading material from this study. Moreover, researchers may conduct a study similar to the present study in a wider scope and that they make a study on the reading variables that has not looked into or on the gaps of the study.

JURNAL REFORMA

Abdullah Farih

Teaching reading a second or foreign language receives a special focus. Some students of foreign language give the most important goal. They get the information through reading, as like newspaper, books, article, etc. good reading text also provide good writing models. So it needs strategies in order to get the best value in the process and the final output. Some strategies are proposed by the expert, but we should select where the appropriate to teach on the students. The aim of the study is to know what the best way to teach reading comprehension students as a second or foreign language is. It is important because nowadays, there are many book sources that use English as their language. This article will discuss (1) the strategies for reading comprehension, (2) the understanding of reading comprehension and teaching reading comprehension. The finding of the study show that the way of teaching reading comprehension is different; and it must adopt the strategy of reading comprehensi...

Premise: Journal of English Education

Junaidi Mistar

This study aimed to assess the student's reading comprehension level and the difficulties that the students faced in reading. The writer conducted this study at Senior High School 2 Pasuruan, especially the 11th-grade students in Language Programs. This study focused on the student's reading ability level and the difficulties that the students faced in reading. The writer used survey research designs in conducting this study. The participants of this study were 30 students. In collecting the data, the writer distributed closed-ended questionnaires, which consisted of 29 questions with a 4-point scale (always, sometimes, rarely, and never). The result showed that 47% of students faced many problems or difficulties understanding the reading text. While 37% of students sometimes found difficulties reading an English passage, 16% of the students who chose "rarely" indicated that the difficulty they faced was not challenging. In conclusion, it was found that many studen...

The Comprehension Problems of Children with Poor Reading Comprehension despite Adequate Decoding: A Meta-Analysis

The purpose of this meta-analysis was to examine the comprehension problems of children who have a specific reading comprehension deficit (SCD), which is characterized by poor reading comprehension despite adequate decoding. The meta-analysis included 86 studies of children with SCD who were assessed in reading comprehension and oral language (vocabulary, listening comprehension, storytelling ability, and semantic and syntactic knowledge). Results indicated that children with SCD had deficits in oral language ( d = −0.78, 95% CI [−0.89, −0.68], but these deficits were not as severe as their deficit in reading comprehension ( d = −2.78, 95% CI [−3.01, −2.54]). When compared to reading comprehension age-matched normal readers, the oral language skills of the two groups were comparable ( d = 0.32, 95% CI [−0.49, 1.14]), which suggests that the oral language weaknesses of children with SCD represent a developmental delay rather than developmental deviance. Theoretical and practical implications of these findings are discussed.

Reading comprehension, or the process of engaging text for the purpose of extracting and constructing meaning ( Snow, 2002 ), has paramount importance to academic success and future life outcomes ( National Institute of Child Health and Human Development [NICHD], 2000 ; Snow, 2002 ). Yet only about 36% of fourth graders and 34% eighth graders in the United States have reading comprehension scores at or above proficiency by the end of the academic year ( U.S. Department of Education, 2015 ). Furthermore, nearly 31% of fourth graders and nearly 24% of eighth graders continue to attain reading comprehension scores that are below even the basic level. This indicates that a substantial proportion of fourth and eighth graders would have problems with more complex activities that extend beyond the text itself (e.g., comparing and contrasting ideas or making inferences beyond the text). This is particularly troubling given the importance of comprehension skills for success in school, in the workplace, and in daily life (e.g., understanding newspapers and forms and contracts to be signed).

Given the importance of decoding to reading comprehension it is not surprising that decoding deficits often result in comprehension difficulties ( Perfetti, 1985 ; Perfetti & Hart, 2001 ; Perfetti & Hogaboam, 1975 ; Perfetti, Landi & Oakhill, 2005 ; Shankweiler et al., 1999 ; Snow, Burns, & Griffin, 1998 ). However, it is estimated that between 10 and 15% of 7- to 8-year-old children have normal performance on decoding measures yet still experience deficits in reading comprehension ( Nation & Snowling, 1997 ; Stothard & Hulme, 1995 ; Yuill & Oakhill, 1991 ); that is, these children are characterized as having a specific reading comprehension deficit (SCD). Although this estimate varies depending on the criteria used to identify children with SCD (see Rønberg & Petersen, 2015 ), large-scale identification studies have shown that the prevalence of SCDis most likely around 8% for children between the ages of 9 and 14 years ( Keenan et al., 2014 ). Even an 8% prevalence rate would mean an average of two students in a classroom could meet the criteria for SCD.

Reading comprehension is a complex process, involving a variety of cognitive and linguistic skills. As a result, deficits in any cognitive ability important to the comprehension process can potentially lead to deficits in reading comprehension performance. Perfetti and colleagues ( Perfetti et al., 2005 ; Perfetti & Stafura, 2014 ) provide a comprehensive framework for understanding the processes and skills involved in reading comprehension; deficits in comprehension could result from a variety of sources beyond decoding, including differences in sensitivity to story structure, inference making, comprehension monitoring, syntactic processing, verbal working memory, and oral language skills ( Cain & Oakhill, 1996 , 1999 ; Cain, Oakhill, Barnes, & Bryant, 2001 ; Nation, Adams, Bowyer-Crane, & Snowling, 1999 ; Nation & Snowling, 1998b , 1999 ; Oakhill, Hartt, & Samols, 2005 ; Pimperton & Nation, 2010a ; Snowling & Hulme, 2012 ).

Existing studies of children with SCD show that they perform poorly on a range of oral language assessments ( Cain, 2003 ; Cain, 2006 ; Cain et al., 2005 ; Cain & Oakhill, 1996 ; Carretti et al., 2014 ; Nation & Snowling, 2000 ; Oakhill et al., 1986 ; Stothard & Hulme, 1996 ; Tong, Deacon, & Cain, 2014 ; Tong, Deacon, Kirby, Cain, & Parrila, 2011 ; Yuill & Oakhill, 1991 ). However, relatively little is known about whether the comprehension problems of children with SCD are the result of their oral language deficits. Although it is possible that the documented deficits in oral language account for the observed deficits in reading comprehension, they may only be a contributing factor. A better understanding of the comprehension problems for children with SCD may be a first step towards better identification and remediation.

We briefly describe relevant theories of reading comprehension because existing theories may inform our understanding of the comprehension problems of children with SCD and understanding the comprehension problems of children with SCD in turn may inform theories of comprehension.

Several theories of reading comprehension have emerged over the years. These include the bottom-up view, the top-down view, the interactive view, the metacognitive view, and the simple view of reading comprehension. Each of these theories are relevant within the present context. Thus, we briefly discuss each theory below.

According to the bottom-up view of reading comprehension, readers move from an understanding of parts of language (e.g., letters, words) to an understanding of meaning or the whole (e.g., phrases, passages; Gough, 1972 ; Holmes, 2009 ; LaBerge & Samuels, 1974 ). Comprehension is thought to be a product of the acquisition of hierarchically arranged subskills ( Dole et al., 1991 ). Thus, lower-level word recognition skills precede the development of more complex skills that lead to an eventual understanding of phrases, sentences, and paragraphs. Automaticity in processing and understanding written text is also thought to affect text comprehension ( LaBerge & Samuels, 1974 ). Automaticity refers to the fact that proficient readers can read text automatically and that they do not need to focus consciously on lower-level word recognition. Thus, children with decoding problems allot greater cognitive resources to word recognition – and less to comprehension – whereas proficient readers are able to devote greater cognitive resources to higher-level cognitive processes (e.g., working memory; Daneman & Carpenter, 1980; Perfetti, 1985 ; Perfetti & Hogaboam, 1975 ).

Based on the top-down (i.e., conceptually-driven) view of reading comprehension, readers are moving from meaning down to the component parts of words as they engage with text ( Rumelhart, 1980 ; Shank & Abelson, 1977 ). According to this view, a reader's mental frameworks or schemas are the driving force behind successful reading comprehension ( Rumelhart, 1980 ). Readers are actively integrating new information that is encountered in the text with information that they have already stored within their previously established mental representations (i.e., background knowledge).

Top-down and bottom-up aspects are combined in the interactive view of reading comprehension. Based on this view, reading comprehension requires the reader to devote attentional resources to the more basic features of the text (e.g., letters, words) while simultaneously focusing on the more general aspects (e.g., syntax, semantics) and actively interpreting what is being read ( Perfetti et al., 2005 ). Proficient readers are those who successfully engage with multiple sources of information provided within the text and information that is not readily available from the text (Kintsch, 1998; Perfetti & Stafura, 2014 ; van Dijk & Kintsche, 1983 ). Good readers are are able to recognize and interact with key features of the text, such as lexical characteristics, at the same time that they are more broadly identifying the purpose of a passage or a paragraph ( Rayner, 1986 ; Rayner et al., 2001 ).

The simple view of reading asserts that reading comprehension is the product of decoding ability and language comprehension ( Gough & Tunmer, 1986 ; Hoover & Gough, 1990 ). The simple view also has substantial empirical validation. For example, decoding has emerged as a reliable predictor of reading comprehension ability in a variety of instances (e.g., Kendeou, van den Broek, White, & Lynch, 2009 ; Shankweiler et al., 1999 ). In fact, poor decoding skills are associated with reading comprehension problems ( Perfetti, 1985 ). Additionally, oral language skills remain a robust and unique predictor of reading comprehension over and above word reading skills ( Nation & Snowling, 2004 ).

Oral language is defined as the ability to comprehend spoken language ( National Early Literacy Panel, 2008 ) and includes a wide variety of skills, such as expressive and receptive vocabulary knowledge, grammar, morphology, syntactic knowledge, conceptual knowledge, and knowledge about narrative structure ( Beck, Perfetti, & McKeown, 1982 ; Bishop & Adams, 1990 ; Bowey, 1986 ; Perfetti, 1985 ; Roth, Speece, & Cooper, 2002 ). Oral language skills impact reading comprehension directly, such as through the understanding of the words presented in a text, as well as indirectly via other literacy-related skills (e.g., phonological awareness; NICHD, 2000 ; Wagner & Torgesen, 1987 ). Furthermore, the unique contribution of oral language to reading comprehension remains even after accounting for word recognition ( Oullette, 2006 ).

The simple view provides a potential explanation for the reading comprehension problems of children with SCD that is consistent with their observed oral language deficits: Reading comprehension requires both adequate decoding and adequate oral language comprehension. This would explain the observation that children with SCD have adequate decoding but not adequate oral language comprehension. Catts, Adolf, and Weismer (2006) and Nation and Norbury (2005) applied this simple view of reading framework to identify different types of reading problems in eighth graders and 8-year-old children, respectively. According to this classification system, children with good decoding and good comprehension are adequate readers; children with poor decoding and poor comprehension are garden-variety poor readers; children with good comprehension and poor decoding meet criteria for dyslexia; and children with good decoding and poor comprehension have SCD. Thus, a mastery of both decoding and language comprehension is necessary for reading proficiency.

Developmental Delay or Developmental Deficit?

Developmental delay and developmental deficit are two hypotheses that are often discussed in relation to the nature of reading disability (e.g., dyslexia; see Francis, Shaywitz, Stuebing, Shaywitz, & Fletcher, 1996 ). The developmental delay hypothesis asserts that poor reading performance results from a delayed acquisition of reading-related skills ( Francis et al., 1996 ). However, these children follow the same developmental trajectory as typical readers ( Francis et al., 1996 ). The developmental deficit hypothesis, on the other hand, states that the underlying skill shows a different or deviant developmental trajectory ( Francis et al., 1996 ). For the case of reading disability, the underlying skill examined was phonological processing. We are interested in determining whether an oral language weakness represents a developmental delay or deficit for children with SCD. This hypothesis could be tested within studies that matched children with SCD to a younger group of typically-developing children (comprehension-age matching; see Cain, Oakhill, & Bryant, 2000 ). If children with SCD demonstrated similar performance to the comprehension-age matched group this would support developmental delay. If children with SCD had worse performance than the comprehension-age matched group, this outcome would support developmental deviance.

The importance of the distinction between developmental delay and developmental deficit is that a skill that is characterized as a developmentally deficit is more likely to be a contributing factor in the development of the reading problem. Developmental delay implies that the skill is consistent with the observed delay in reading and is therefore less likely to be a contributing factor. To our knowledge, an empirical examination of these two hypotheses has not yet been conducted for the observed oral language deficits in children with SCD.

Below, we describe a study conducted by Cain and Oakhill (2006) that has several characteristics that are typical of studies involving children with SCD. In this investigation, the authors were interested in the cognitive profiles of 7- to 8-year-old children with SCD; this age range is very common for investigations of children with SCD (e.g., Cain, 2003 ; Cain & Oakhill, 1996 , 2007; Jerman, 2007; Oakhill, 1982 ). Children were selected based on their performance on measures of reading comprehension and word reading accuracy and were followed longitudinally. In this case, the Neale Analysis of Word Reading Ability was used to categorize children into groups of good and poor comprehenders. Age-appropriate word reading accuracy was defined as being between 6 (lower limit) and 12 months (upper limit) of their chronological age (e.g., Clarke, 2009 ). Poor reading comprehension was defined as a 12-month discrepancy between comprehension age and chronological age and their reading accuracy age and comprehension age (e.g., Nation & Snowling, 1999 , 2000 ; Weekes, Hamilton, Oakhill, & Holliday, 2008 ). Typical readers are defined as attaining reading comprehension scores that are at or above word reading accuracy performance. Due to one-to-one matching and the low proportion of SCD in the population, final groups were small (23 children per group); this is typical of many studies involving children with SCD (e.g., Ehrlich & Remond, 1997 ; Geva & Massey-Garrison, 2012 ; Nation & Snowling, 1998a , 1998b ). In this study, children were given a battery of assessments that included a combination of standardized and experimenter-created measures (e.g., Nation et al., 1999 ; Nation & Snowling, 2000 ). A unique aspect of this investigation is that children were followed longitudinally; many studies involving children with SCD are single time point studies (e.g., Cain & Oakhill, 1999 ; Oakhill, 1983 ).

SCD has been defined in a variety of ways across different studies. Although researchers tend to agree on the need for a discrepancy between an individual's decoding ability and their reading comprehension skills, individuals with SCD (also referred to as poor comprehenders or less-skilled comprehenders in the literature) have been identified using one of four criteria:

• A discrepancy between reading comprehension and decoding (e.g., Isakson & Miller, 1976 ; Nation & Snowling, 1998a ; Oakhill, Yuill, & Parkin, 1986 ; Pimperton & Nation, 2010a );
• A discrepancy between reading comprehension and decoding with an additional requirement that decoding skills are within the normal range (e.g., Cain et al., 2001 ; Cataldo & Oakhill, 2000 ; Cragg & Nation, 2006 ; Torppa et al., 2009);
• Discrepancies between reading comprehension, decoding, and chronological age with an additional requirement that decoding skills are within the normal range ( Cain, 2003 ; Cain, 2006 ; Cain et al., 2000 ; Cain & Oakhill, 2006 , 2011 ; Cain, Oakhill, & Elbro, 2003 ; Cain, Oakhill, & Lemmon, 2004 ; Cain & Towse, 2008 ; Clarke, 2009 ; Marshall & Nation, 2003 ; Nation & Snowling, 1997 , 2000 ; Nation et al., 2001 ; Oakhill et al., 2005 ; Spooner, Gathercole, & Baddeley 2006 ; Stothard & Hulme, 1995 ; Yuill, 2009 ; Yuill & Oakhill, 1991 );
• A discrepancy between reading comprehension and word-level decoding with additional requirements that decoding skills are within the normal range and that comprehension scores fall below a given percentile or cut point ( Cain & Towse, 2008 ; Carretti, Motta, & Re, 2014 ; Catts et al., 2006 ; Henderson, Snowling, & Clarke, 2013 ; Kasperski & Katzir, 2012; Megherbi, Seigneuric, & Ehrlich, 2006 ; Nation, Clarke, Marshall, & Durand, 2004 ; Nation, Snowling & Clark, 2007 ; Nesi, Levorato, Roch & Cacciari, 2006 ; Pelegrina, Capodieci, Carretti, & Cornoldi, 2014 ; Pimperton & Nation, 2014 ; Ricketts, Nation, & Bishop, 2007 ; Shankweiler et al., 1999 ; Tong et al., 2011 ; Tong et al., 2014 ).

Despite the fact that differences in identification criteria influence the percentage of children identified as having SCD (see Rønberg & Petersen, 2015 ), children with SCD likely represent a small but significant proportion of struggling readers. Moreover, across studies included within the present review, SCD was identified using all of these different criteria. Therefore, our findings provide an overall estimate of the nature of children's comprehension problems regardless of identification method.

The purpose of the present meta-analysis is to better understand the comprehension deficits of children who have SCD. The framework for the present meta-analysis grew out of a recent investigation that tested three hypotheses regarding the nature of the comprehension problem in a large sample of over 425,000 first-, second-, and third graders with SCD ( Spencer, Quinn, & Wagner, 2014 ). The three hypotheses tested whether comprehension problems for these children were largely specific to reading, general to oral language, or both (i.e., a mixture). Children were obtained from a statewide database, and prevalence of SCD was calculated based on percentile cutoffs. The results indicated that over 99 percent of children in each grade who had SCD also had deficits in vocabulary knowledge. This finding indicates that children's comprehension deficits were general to reading and at least one important aspect of oral language.

Although these results provide compelling evidence that comprehension problems are general to at least one aspect of oral language (i.e., vocabulary knowledge), three limitations of the study need to be noted. First, participants included mostly children attending Reading First schools, a Federal program for improving reading performance for students from low socioeconomic backgrounds. Because poverty is a risk factor for delayed development of oral language, the results may not generalize to students not living in poverty. Second, the assessments were brief and receptive vocabulary knowledge served as the only measure of oral language comprehension, when in fact, oral language is potentially comprised of a variety of different skills that might affect reading comprehension. Third, the study did not compare the relative magnitudes of the deficits observed in reading comprehension and vocabulary, a potentially important new source of data that could be used to compare alternative hypotheses about the nature of the comprehension problems of children with SCD.

These limitations suggest the need for a comprehensive review of the literature on the nature of the comprehension problems of children who have SCD. Such a review could incorporate results from studies with more representative samples and using a variety of measures. By examining magnitudes as well as the existence of deficits in reading versus oral-language comprehension, it would be possible to test a previously neglected hypothesis in Spencer et al. (2014) , namely that children with SCD could have deficits in oral language that are not as severe as their deficits in reading comprehension.

Thus, in addition to testing two hypotheses from Spencer et al. (2014) – (a) Children with SCD have comprehension deficits are specific to reading, such that they demonstrate impaired reading comprehension but no impairments in oral language and (b) children with SCD have comprehension deficits are general to reading and oral language, such that they demonstrate equal impairment in reading comprehension and oral language – we also test a third hypothesis in the present meta-analysis, (c) children with SCD have comprehension deficits that extend beyond reading to oral language, but they demonstrate greater impairment in reading comprehension than in oral language.

Hypothesis one: children with SCD have comprehension problems that are specific to reading

Theoretical support for this hypothesis comes from the bottom-up view of reading comprehension and from the automaticity of reading ( Gough, 1972 ; Holmes, 2009 ; LaBerge & Samuels, 1974 ). It is possible that children might have adequate decoding but their adequate decoding requires processing resources that are then not available for comprehension while reading. If this were the case, their comprehension would be impaired for reading comprehension because decoding is required but not impaired for oral language.

Empirical support for this hypothesis comes from studies that demonstrate the existence of individuals who have been identified as having SCD in the presence of intact or relatively intact vocabulary knowledge ( Cain, 2006 ; Nation, et al., 2010 ). Moreover, some studies that compared children with and without SCD matched them on vocabulary performance (e.g., Cain, 2003 , Cain, 2006 ; Spooner et al., 2006 ; Tong et al., 2014 ). That it was possible to do this match supports the possibility that comprehension problems are specific to the domain of reading.

Hypothesis two: children with SCD have comprehension problems that are general to reading and oral language

Several theoretical perspectives provide a rationale for this hypothesis, including the simple view, top-down view, and interactive views of reading comprehension. The simple view ( Gough & Tunmer, 1986 ; Hoover & Gough, 1990 ) provides support for this hypothesis because it explains SCD as resulting from a deficit in oral language comprehension ( Catts et al., 2006 ; Nation et al., 2004 ). The top-down and interactive views are in line with this hypothesis because both frameworks emphasize the readers' mental frameworks ( Rumelhart, 1980 ; Shank & Abelson, 1977 ). The top down processing highlighted in both frameworks would affect comprehension regardless of whether the context is written or oral language.

Empirical support for this hypothesis comes from studies showing that oral language ability is a predictor of future reading comprehension success and failure ( Nation & Snowling, 2004 ; Snow et al., 1998 ); children with reading comprehension problems tend to have deficits in oral language ( Catts, Fey, Tomblin, & Zhang, 2002 ). For example, Catts, Fey, Zhang, and Tomblin (1999) investigated relations between oral language and reading comprehension skills in second graders. Results indicated that children with reading comprehension deficits were significantly more likely to have had oral language weaknesses in kindergarten compared to students with more typical comprehension development (see also Elwer, Keenan, Olson, Byrne, & Samuelsson, 2013 ).

The view that comprehension problems are general to oral language and reading is supported by multiple investigations. Children with SCD have demonstrated weaknesses related to a variety of oral language domains, such as semantic processing, listening comprehension, and syntactic ability ( Carretti, Motta, & Re, 2014 ; Nation & Snowling, 2000 ; see Cain & Oakhill, 2011 and Justice, Mashburn, & Petscher, 2013 for longitudinal evidence). When compared to typical readers, these children also tend to perform significantly poorer on measures tapping verbal working memory skills (see Carretti, Borella, Cornoldi, & De Beni, 2009 ). Differences between typically-developing readers and individuals with SCD have also been reported using a wide variety of behavioral and EEG/ERP measures (e.g., Landi & Perfetti, 2007 ).

Hypothesis three: children with SCD have comprehension problems that extend to oral language but are less severe for oral language than for reading

Theoretical support for this hypothesis is provided by a combination of theoretical rationales discussed for the previous two hypotheses. Specifically, a deficit that is general to oral language as well as reading comprehension is assumed, combined with additional deficits that are specific to reading. For example, a deficit in vocabulary would impair performance in reading comprehension and oral language. Simultaneously, decoding and orthographic processing could require attention and cognitive resources that are not required by listening, such as visual processing. The combined result would be impairments in both oral language and reading comprehension, but the impairment would be greater for reading comprehension.

Empirical support for this hypothesis comes from studies showing that these children demonstrate differential performance across various oral language tasks ( Cain, 2003 ; Cain, 2006 ; Cain, Oakhill, & Lemmon, 2005 ; Stothard & Hulme, 1992 ; Tong et al., 2014 ). For example, Cain (2003) examined language and literacy skills in children with SCD who were matched to typical readers based on vocabulary; however, these same children exhibited significantly poorer performance on other oral language tasks, such as listening comprehension and a story structure task. Similarly, Tong et al. (2014) included children with SCD who were vocabulary-matched to typical readers. Yet, children with SCD exhibited poor performance on a morphological awareness task. Therefore, it may be that the comprehension problems of children with SCD affects some but not all aspects of oral language.

Additionally, we were interested in examining the effect of several potential moderators of effect size outcomes, specifically the effects of (a) publication type, (b) participant age, and (c) type of oral language measure. The rationale for these moderators are as follows: First, if publication type (e.g., published journal article versus unpublished dissertation) significantly predicts effect size outcomes, we would attribute this, at least partially, to publication bias. Thus, we wanted to include this variable within each meta-analysis. Second, we were interested in participant age as a moderator of effect sizes ( Catts et al., 2006 ; Elwer et al., 2013 ; Nation, Cocksey, Taylor, & Bishop, 2010 ). Previous research has also indicated that younger children with SCD tend to have weaker reading comprehension skills compared to older children ( Authors, 2017 ). We sought to investigate whether this finding would be replicated within a different sample and also whether these differences transfer to oral language skills as well. Finally, type of oral language measure was included as a potential moderator due to the fact that oral language measures vary greatly in the skills that they assess ( Cain & Oakhill, 1999 ; Nation et al., 2004 , 2010 ; Tong et al., 2011 ). For instance, a receptive vocabulary assessment is likely to be much less difficult for a child with SCD compared with a syntactic or morphological task. Therefore, examining the potential effects of type of oral language measure may provide additional insight into which tasks may be best to use for identifying children with SCD.

Across four decades, multiple systematic reviews of reading comprehension have been conducted. These reviews have examined a variety of topics, including an examination the component skills of reading comprehension and intervention research for struggling readers (e.g., Bus & van Ijzendoorn, 1999 ; Ehri, Nunes, Stahl, & Willows, 2001 ; Swanson, Tranin, Necoechea, & Hammill, 2003 ). In more recent years, there have been several narrative reviews focusing specifically on children with SCD ( Hulme & Snowling, 2011 ; Nation & Norbury, 2005 ; Oakhill, 1993 ), but only one known meta-analysis to date has investigated the cognitive skills of these individuals ( Carretti et al., 2009 ). However, Carretti et al. (2009) focused exclusively on working memory skills whereas the present investigation examines performance of children with and without SCD on a wide array of oral language tasks in addition to verbal working memory.

In the present review, we examine studies using five methods. First, we conducted between-group meta-analyses comparing the reading comprehension performance of children with SCD with the reading comprehension performance of typically-developing readers. Second, we conducted between-group analyses comparing the oral language performance (as indexed by measures of vocabulary, listening comprehension, storytelling ability, morphological awareness, and semantic and syntactic knowledge) of children with SCD with the oral language performance of typically-developing readers. Third and fourth, we conducted the same meta-analyses for reading comprehension and oral language performance for studies that included a comprehension-age matched group (see Cain et al., 2000 ). The existence of such studies makes it possible to determine whether impaired oral language performance represents developmental delay (i.e., performance similar to younger normal comprehenders) or a developmental difference (i.e., performance different than that of younger normal comprehenders; Francis et al., 1996 ). Finally, we conducted a separate meta-analysis for studies reporting performance on standardized reading comprehension and oral language measures for the same participants (i.e., a within-child comparison of reading comprehension and oral language) because we were interested in the comparability of oral language skills to reading comprehension within children who have SCD.

Study Collection

The current meta-analysis includes studies published in English from January 1, 1970 to February 20, 2016. Several electronic databases and keywords were used to locate relevant studies. These databases included PsycINFO, ERIC, Medline, and ProQuest Dissertations. In an effort to reduce the likelihood of publication bias within the present review, we also searched several gray literature databases (i.e., SIGLE, ESRC, and Web of Science ). We used title-based keywords related to reading comprehension and reading disabilities ( specific comprehension deficit*, poor comprehender*, comprehension difficult*, less-skilled comprehen*, comprehension failure, reading difficult*, difficulty comprehending, poor comprehension, struggling reader*, specific reading comprehension difficult*, specific reading comprehension disabilit*, low comprehender*, weak reading comprehen*, reading comprehension disab*, poor reading comprehension ) in combination with other reading-related keywords ( reader*, reading, subtype*, subgroup ). Our search spanned peer-reviewed and non-peer-reviewed journal articles, dissertations and theses, book chapters, reports, and conference proceedings. The references of relevant articles were also hand searched, and we contacted researchers who had at least three relevant publications (first authored or not) as a way of including unpublished data within the present review. We conducted additional searches for these same researchers using author- and abstract-based keyword searches [au( author ) AND ab( comprehen* )].

Inclusionary criteria

Several inclusionary criteria were used to select studies to be included within the present synthesis. Studies were required to: (a) report original data (i.e., sample means, standard deviations, correlations, sample sizes, t -tests, and/or F -tests); (b) include native speakers of a language; (c) assess children between the ages of 4 and 12 years; (d) contain at least one measure of reading comprehension, decoding ability, and oral language; (e) include a sample of children with SCD based on their performance on measures of reading comprehension and decoding ability; and (f) include a typically-developing group of readers for comparisons 2 .

We applied the language-based criterion because we wanted to be able to investigate the relation between poor reading comprehension and oral language skills separate from language status because language status is known to affect reading comprehension (e.g., Kieffer, 2008 ). However, studies could include monolingual samples that spoke a language other than English (e.g., Italian) provided that the study was reported in English. Acceptable measures of reading comprehension included assessments that measured individuals' comprehension of the text beyond word reading ability; acceptable measures of decoding ability included assessments that measured real word decoding, nonword decoding, and/or reading accuracy; and acceptable measures of oral language included tasks that assessed vocabulary knowledge, syntactic and semantic processing, listening comprehension, and/or storytelling ability.

Exclusionary criteria

Three exclusionary criteria were applied for studies included in the current meta-analysis: (a) teacher and parent ratings were not acceptable methods for identifying children with SCD, (b) samples of non-native speakers, and (c) samples could not also contain children characterized as having intellectual disability, attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), aphasia, hydrocephalus, or hearing or vision impairments.

Final study selection

The initial search yielded approximately 3,050 results. After eliminating duplicates, studies that did not adhere to our inclusion/exclusion criteria, and studies reporting results from identical participants, a total of 86 studies remained.

A random sample of 10% of the studies was coded twice by the first author and a graduate student in order to establish inter-coder reliability; studies were coded based on study features (i.e., study type, sample size, operational definition of SCD, matching variables, language spoken, and sample age) and reading comprehension- and oral language-related constructs (i.e., reported reliabilities, correlations with oral language measures, means and standard deviations for each assessment, and reported t values or F ratios). We additionally coded participant age, type of oral language measure (i.e., vocabulary knowledge, narrative, listening comprehension, syntactic/grammar, semantic knowledge, and figurative language), and type of publication (i.e., journal article, book chapter, theses/dissertations, and unpublished data). Cohen's kappa was used to measure inter-coder reliability (96% for study features; 98% for reading comprehension-related constructs; 94% for the oral language-related constructs). The overall reliability exceeded acceptability of kappa ≥ .70 (kappa = 96%). Discrepancies were resolved through discussion or by referring to the article.

The final sample included 84 studies for between ( k brc = 152 effect sizes for reading comprehension ; k bol = 309 effect sizes for oral language) and within-group analyses ( k wrc = 97 effect sizes). The between-group analyses were twofold. One was a comparison of children with SCD to typical readers and another was a comparison of children with SCD to a comprehension-age matched group of children. Between-group comparisons of children with SCD to typical readers allowed for a test of the three hypotheses outlined previously: (a) children with SCD have comprehension problems that are specific to reading; (b) children with SCD have comprehension problems that are general to reading and oral language; or (c) children with SCD have comprehension problems that extend to oral language but are less severe for oral language than for reading. Between-group comparisons of children with SCD to a comprehension-age matched group allowed for a test of the delay versus deficit hypotheses for the anticipated oral language difficulties. A subsample of the original study sample ( n = 4) included comprehension-age matched groups for additional analyses ( k brc = 4 effect sizes for reading comprehension ; k bol = 30 effect sizes for oral language).

Within-child analyses require that both measures within a single study use the same scale. Thus, in order to be included within the within-child analysis, studies had to include standardized measures of reading comprehension and oral language and report standard scores, scaled scores, z -scores, or t -scores. Our within-child analyses allowed us to test the robustness of the pattern of results observed in the between-group comparison. That is, we were able to compare the reading comprehension and oral language skills within children who had SCD.

Meta-Analytic Methods

All analyses were conducted using Microsoft Excel (Version 14.0), and Metafor ( Viechtbauer, 2010 ) and Robumeta packages in R ( Fisher & Tipton, 2015 ). Effect sizes were calculated using Hedge's g (Hedges, 1981), which is Cohen's d ( Cohen, 1977 ) after incorporating a correction for small sample sizes. Negative effect size values indicate that children with SCD had a lower group mean than typically developing readers. In several instances, groups were vocabulary-matched (i.e., children with SCD were selected on the basis of having average vocabulary performance compared to a group of typical readers). 3

Average weighted effect sizes for each meta-analysis were calculated using random-effects models, which assume all parameters to be random as opposed to fixed ( Shadish & Haddock, 2013 ). We used random-effects models in the present investigation because Q (i.e., homogeneity of effect size; Hedges & Olkin, 1985 ) was rejected across most comparisons. For one comparison, Q was not rejected; for this meta-analysis, we used a fixed-effects model. We also estimated I 2 , which calculates the percentage of variance due to heterogeneity. We used random-effects models to calculate a 95% confidence interval (CI) in order to determine whether each calculated average weighted effect size was statistically significant (i.e., different from zero). A CI within random-effects models assumes systematic study variability (i.e., that differences across studies do not result from random sampling error; Shadish & Haddock, 2013 ). We additionally conducted an Egger test for funnel plot asymmetry within each meta-analysis as a means of testing whether publication bias was present (significant plot asymmetry) or absent (non-significant plot asymmetry; Egger, Smith, Schneider, & Minder, 1997 ).

Across meta-analyses, there were several instances in which a single study resulted in multiple effect size estimates. We used robust variance estimation with the small sample size correction to handle dependent effect sizes ( Hedges, Tipton, & Johnson, 2010 ; Tipton, 2015 ). This relatively recent approach has advantages over alternative approaches to handling dependent effect sizes such as including only one effect size per study, creating an average effect size, or using multivariate approaches to model the dependency. Robust variance estimation allows one to use all effect sizes including multiple ones from the same sample in the meta-analysis for estimating average weighted effect sizes and for testing possible moderators, then corrects for the effects of the dependencies in the significance testing. Although robust variance estimation can be implemented in macros to common statistical packages such as SPSS, an efficient way of doing so is by using the Robumeta package available in R ( Fisher & Tipton, 2015 ). We carried out meta-regressions analyses of potential moderators using Robumeta when there were dependent effect sizes. For meta-analyses that did not demonstrate dependency among effect size estimates (i.e., between group comparison of reading comprehension for children with SCD and comprehension-age matched children), we calculated the average weighted effect size estimate using traditional methods in Metafor.

A total of 86 independent studies were included within the analyses. Effect sizes for each comparison are reported in Table 1 (see also Appendices A, B, and C). A substantial portion of studies included English-speaking samples (Study n = 72). Fourteen studies included children who spoke Italian ( n = 5), French ( n = 3), Finnish ( n = 1), Hebrew ( n = 1), Chinese ( n = 2), Portuguese ( n = 1), and Spanish ( n = 1). Across studies, children were between the ages of 4 and 12 years.

Note. k = Number of effect sizes; d = Average-weighted effect size estimate; CI = Confidence interval; SCD = Children with specific reading comprehension deficits;

Effect Size Analyses

Comparisons of children with scd to typical readers.

We compared children with SCD to typical readers on measures of reading comprehension and oral language. These analyses served as a means to test whether: (a) children with SCD have comprehension problems that are specific to reading; (b) children with SCD have comprehension problems that are general to reading and oral language; or (c) children with SCD have comprehension problems that extend to oral language but are less severe for oral language than for reading.

Reading comprehension

One hundred and fifty-two comparisons were made for the reading comprehension of children with SCD and typically-developing readers (Study n = 84). Across studies, there were 17,600 children with SCD ( M = 209.53; SD = 703.14; range: 7-3,236) who were compared with 155,874 typically developing children ( M = 1,855.64; SD = 6,737.96; range: 8-29,676). The average weighted effect size was negative, large, and statistically significant (random-effects robust variance estimation: d = −2.78, 95%CI [−3.01, −2.54]). Because the CI does not include zero, this indicates that the effect size estimate is significantly different from zero. This suggests that children with SCD performed substantially poorer on measures of reading comprehension compared to their typically developing peers, which was expected. Study-specific effect sizes for reading comprehension, participant ages, and sample sizes for these comparisons are reported in Appendix A ; effect sizes are reported in descending order. There was a large variability in effect size estimates across studies due to heterogeneity, I 2 = 94.39 (see Table 1 ). Sensitivity analyses indicated that varying values of rho (ρ) from 0 to 1 in .20 increments did not affect tau squared (τ 2 ), the subsequent weights, and the average weighted effect size estimate. This outcome suggests that the observed effect size is fairly robust. An Egger test of funnel plot asymmetry was significant, z = −7.09, p < .0001 (see Figure 1 ), indicating asymmetry in effect size estimates across studies.

Funnel plots for between- and within-group comparisons. Note . RC = Reading comprehension; OL = Oral language; WM = Working memory; CAM = Reading-comprehension age-match.

Oral language

Three hundred and nine comparisons were made for the oral language skills of children with SCD and typically-developing children, (Study n = 76). There were 16,494 children with SCD ( M = 219.93; SD = 706.39; range: 7-3,016) who were compared with 144,857 typically developing children ( M = 1,931.43; SD = 6,676.47; range: 8-28,970). The average weighted effect size was also negative, large, and statistically significant (random-effects robust variance estimation: d = −0.78, 95% CI [−0.89, −0.68]). Thus, when compared to children without comprehension problems, children with SCD additionally exhibit difficulty completing oral language tasks; however this deficit was not as severe as for reading comprehension. Study-specific effect sizes for oral language, participant ages, and sample sizes for these comparisons are reported in Appendix A ; effect sizes are reported in descending order. Variability due to heterogeneity was large across studies, I 2 = 85.55 (see Table 1 ). Sensitivity analyses indicated that the observed effect size is fairly robust; varying values of ρ resulted in no differences. An Egger test of funnel plot asymmetry was significant, z = −2.11, p < .05 (see Figure 1 ), suggesting some asymmetry in estimates. Additionally, we also examined verbal working memory for studies that were already included in the analysis, which added 91 additional comparisons to the analysis. The average weighted effect size remained negative, large, and statistically significant (random-effects robust variance estimation: d = −0.77, 95% CI [−0.87, −0.67]; I 2 = 85.12; see Table 1 ).

It is important to note that across comparisons of reading comprehension and oral language, different studies were available for analyses; however, when we analyzed only overlapping studies (Study n = 74), the effects for reading comprehension (random-effects robust variance estimation: d = −2.80, 95% CI [−3.05, −2.55]; I 2 = 94.68) and oral language were nearly identical (random effects robust variance estimation: d = −0.79, 95% CI [−0.90, −0.68]; I 2 = 85.50).

Comparisons of children with SCD to comprehension-age matched readers

Given that we found evidence that children with SCD do exhibit deficits in oral language, we were additionally interested in how such deficits were best characterized. Thus, we conducted a between-groups meta-analysis that compared the performance of children with SCD to younger comprehension-age matched readers. Children in the comprehension-age matched group were selected on the basis of having performance equivalent to children with SCD (see Cain et al., 2000 ). 4 Across studies, children within the comprehension-age matched group were approximately two years younger than children with SCD.

Four comparisons were made for the reading comprehension skills of children with SCD and comprehension-age matched control children (Study n = 4). There were 73 children with SCD ( M = 18.25; SD = 7.23; range: 14-29) compared with 68 typically-developing children across studies ( M = 17.00; SD = 6.78; range: 14-27). Study-specific effect sizes for reading comprehension, participant ages, and sample sizes for these comparisons are reported in Appendix B ; effect sizes are reported in descending order. The average weighted effect size was moderate and negative, but it was not statistically significant (fixed-effects: d = −0.31, 95% CI [−0.31, 0.02)]; Q (3) = .38, p = .94, I 2 = <1%; see Table 1 ). This outcome was expected given that the two groups were matched for reading comprehension performance. An Egger test of funnel plot asymmetry was non-significant, z = −.13, p = .90 (see Figure 1 ).

Thirty comparisons were made for the oral language skills of children with SCD and children within comprehension-age matched groups (Study n = 4). There were 73 children with SCD ( M = 18.25; SD = 7.23; range: 14-29) and 68 typically-developing children across studies ( M = 17.00; SD = 6.78; range: 14-27). The average weighted effect size was moderate and in favor of comprehension age-matched readers, but it was not statistically significant (random-effects robust variance estimation: d = 0.32, 95% CI [−0.49, 1.14]). These findings suggest that the oral language performance of children with SCD is similar to the performance of younger typical readers. In other words, there is a developmental delay in the oral language skills of children with SCD. Study-specific effect sizes for oral language, participant ages, and sample sizes for these comparisons are reported in Appendix B ; effect sizes are reported in descending order. Across studies, the variability due to heterogeneity was relatively high, I 2 = 77.13 (see Table 1 ). Sensitivity analyses indicated that the observed effect size was quite robust; varying values of ρ resulted in a .02 difference (τ 2 =.402 when ρ = 0; .423 when ρ = 1), which was minimal. However, because the degrees of freedom for these analyses were less than four, it is important to interpret these results cautiously ( Fisher & Tipton, 2015 ). An Egger test of funnel plot asymmetry was non-significant, z = −0.71, p = .48 (see Figure 1 ).

Within-child comparisons of reading comprehension and oral language for children with SCD

In addition to comparing the language and literacy skills of children with SCD to typically-developing readers and comprehension age-matched readers, we also compared the oral language skills to reading comprehension within children who have SCD. The aim of this meta-analysis was so test to robustness of the results (i.e., would the same pattern of findings emerge if comparisons were made for the same group of children [within-group] as opposed to comparisons across different groups [between-group]). Thus, we additionally conducted analyses that examined the reading comprehension and oral language skills within individuals.

Ninety-seven comparisons were included within the analysis (Study n = 32). There were 12,711 children with SCD ( M = 397.22; SD = 822.21; range:7-2,830). Because these analyses included children with SCD, we corrected correlations for range restriction using Thorndike's (1949) correction equation. 5 The average weighted effect size was moderate, negative, and statistically significant (random-effects robust variance estimation: d = −0.84, 95% CI [−1.06, −0.62]), which indicates that the reading comprehension skills of children with SCD are significantly weaker than their oral language skills. These results can be found in Table 1 . Study-specific effect sizes, participant ages, and sample sizes for these comparisons are reported in Appendix C ; effect sizes are reported in descending order. Across studies, the variability due to heterogeneity was substantial, I 2 = 96.06. However, sensitivity analyses indicated that the observed effect size was fairly robust; varying values of ρ resulted in no difference in estimates of τ 2 . An Egger test of funnel plot asymmetry was non-significant for these comparisons, z = 1.33, p = .18 (see Figure 1 ).

It is important to note that different sets of studies were included within our analyses of between-group and within-child comparisons. This may explain why the difference between reading comprehension and oral language performance within children ( d = −0.84) was not equivalent to the differences found between groups for reading comprehension and oral language (effect size difference between −2.78 and −0.78 was −2.00). We empirically tested this by analyzing only those studies that were included within the between-group reading comprehension (random-effects robust variance estimation: d = −2.73, 95% CI [−3.05, −2.42]; I 2 = 96.82) and oral language comparisons (random-effects robust variance estimation: d = −0.95, 95% [CI −1.06, −0.83]; I 2 = 91.00) and the within-child comparisons. Applying this method, we achieved anoticeable reduction in the effect size differences across comparisons (effect size difference between −2.73 and −0.95 was 1.78). This outcome may be a partially due to the absence of publication bias within the within-group comparisons relative to the potential presence of publication bias within the reading comprehension and oral language comparisons.

Moderator Analyses

Metaregressions of study type, age, and oral language measures for comparisons of children with scd to typical readers.

Due to the substantial amount of heterogeneity across studies, we were interested in examining three possible moderators – age, type of oral language measure, and study type (i.e., published journal article, book chapter, thesis/dissertation, unpublished data) – that may explain effect size differences among various studies (see Table 1 and Appendices D and E ). Due to the dependency of effect sizes across studies, we used robust variance estimation to conduct moderator analyses for the present comparisons.

Study type, β = .14, p > .05, t (11.8) = 1.05, was not a significant moderator of differences in effect size estimates for reading comprehension for comparisons of children with SCD to typical readers. However, age, β = −.47, p < .05, t (23.9) = −2.53, was a significant moderator of effect size differences. Next, we examined moderators for comparisons of oral language. Neither study type nor age were significant moderators of differences in effect size outcomes for oral language, β = −.04, p > .05, t (17) = −0.77 for study; β = −.06, p > .05, t (20.1) = −0.85 for age. Because oral language was assessed using different measures across studies, we also conducted a metaregression to examine the potential for differences in oral language measures to be a moderator of effect size outcomes. Because oral language varied both within and across studies, it is important to include both the mean (i.e., between-study covariate) and mean-centered predictors (i.e., within-study covariate) within the moderator analyses to account for the potentially hierarchical structure of the effect size dependencies ( Fisher & Tipton, 2015 ). Using this method, type of oral language measure was not a significant moderator of effect size across studies, β m = −.05, p > .05, t (16.5) = −0.91; β mc = .00, p > .05, t (16.9) = 0.02.

Metaregressions of study type, age, and oral language measures for comparisons of children with SCD to comprehension-age matched readers

We also examined potential moderators within our reading comprehension age-matched comparisons (see Table 2 ). Similar to our between group comparisons, the type of oral language measure, β m = −.10, p > .05, t (1.08) = −0.18; β mc = −.23, p > .05, t (1.20) = −1.05, was not a significant moderator of effect size for the oral language comparisons. 6 However, because the degrees of freedom were less than four, this finding should be interpreted cautiously. Study type and the age range of participants was constant across studies, thus negating the need to conduct moderator analyses for these constructs for the reading comprehension and oral language comparisons.

Note . SCD = Children with specific reading comprehension deficits;

Metaregressions of study type, age, and oral language measures for within-child comparisons

We examined the moderators of study type, age, and oral language measure within our within-group comparisons as well, which are summarized in Table 2 . Study type was a significant predictor of differences in effect size, β = −.24, p < .01, t (15.3) = −2.77. Similarly, type of oral language measure was a significant predictor at the mean, β m = .20, p < .01, t (15.40) = 2.35; β mc = −.03, p > .05, t (8.30) = −0.85. Age, however, was a non-significant predictor in the model, β = −.00, p > .05, t (12.9) = −0.02.

The aim of the present meta-analysis was to determine the nature of the comprehension problems for children with SCD. This investigation was guided by three competing hypotheses: (a) children with SCD have comprehension deficits that are specific to reading; (b) children with SCD have comprehension deficits that are general to reading and oral language; or (c) children with SCD have comprehension problems that extend beyond reading but are more severe for reading than for oral language. The findings of the present meta-analysis support the third hypothesis. Children's weakness in oral language was substantial ( d = −0.78), but not as severe as their deficit in reading comprehension ( d = −2.78). The effects size estimates for oral language were comparable regardless of whether verbal working memory was included in the analysis ( d = −0.77). Within-child comparisons also indicated that performance in reading comprehension was worse than for oral language ( d = −0.84). The pattern of poorer performance in reading comprehension compared to oral language was consistent across all analyses.

When compared to comprehension age-matched readers, children with SCD tended to have comparable oral language ( d = 0.32, ns ) and reading comprehension skills ( d = −0.31, ns ). The fact that older children with SCD did not differ from younger normal readers on reading comprehension was expected rather than informative because the groups were matched on reading comprehension. However, the fact that they did not differ in oral language is informative. It supports the idea that the oral language weaknesses for children with SCD are best characterized as arising from a developmental delay as opposed to a developmental deviance ( Francis et al., 1996 ). A developmental deviance would have been supported had the oral language performance of the older children with SCD been worse than that of the younger comprehension-age matched normal readers.

Overall, our results are consistent with previous investigations. Children with SCD perform poorly on a range of oral language assessments including receptive and expressive vocabulary knowledge, listening comprehension, story structure, knowledge of idioms, awareness of syntactic structure, and morphological awareness among others ( Cain, 2003 ; Cain, 2006 ; Cain & Oakhill, 1996 ; Cain et al., 2005 ; Carretti et al., 2014 ; Nation & Snowling, 2000 ; Oakhill et al., 1986 ; Stothard & Hulme, 1996 ; Tong et al., 2011 , 2014 ; Yuill & Oakhill, 1991 ). These weaknesses emerged despite children's adequate decoding and seemingly intact phonological processing abilities ( Nation & Snowling, 2000 ; Nation et al., 2007 ; Stothard & Hulme, 1992 ). Yet, this pattern makes sense given that phonological processing appears to underlie decoding ability ( Nation et al., 2007 ; Shankweiler et al., 1999 ; Stothard & Hulme, 1996 ).

Explanations for Greater Deficits in Reading Comprehension than in Oral Language

A number of possible explanations for the observed discrepancies between reading comprehension and oral language exist. Although it is not possible to test alternative explanations in the context of the present meta-analysis, they could be tested in future studies.

A latent decoding deficit

At first glance, it seems counterintuitive that a decoding deficit would explain comprehension differences in children with SCD. However, in several studies, only decoding accuracy was used to categorize children (e.g., Cain & Oakhill, 2006 ). It is possible to be adequate in decoding accuracy yet inadequate in decoding fluency. In fact, this is a common outcome of intervention studies (e.g., de Jong & van der Leij, 2003; Torgesen & Hudson, 2006 ). The effortful application of phonics rules or other decoding strategies can result in accurate but slow decoding. This could impair reading comprehension because children's reading would be less automatic ( LaBerge & Samuels, 1974 ) and/or because fewer cognitive resources would be available for comprehension (e.g., Perfetti, 1985 ). This possible explanation could be tested in future studies by using measures of decoding fluency as well as accuracy. A dual-task paradigm could also be used to determine whether the cognitive resources required by decoding were comparable for children with and without SCD.

Differences between written and oral language

Written language differs from oral language in important ways ( Perfetti et al., 2005 ). Written language oftentimes contains more complex sentence structures and more difficult vocabulary than spoken language ( Akinnaso, 1982 ; Halliday, 1989 ). Thus, if children are having difficulty completing tasks that require the use of syntactic knowledge, for instance, they will most likely have difficulty reading grammatically complex texts. Fundamental differences between written and spoken text may also extend to increased demands on background knowledge (e.g., Wolfe & Woodwyck, 2010 ). Background knowledge has been identified as a critical component within several models of reading comprehension ( Kintsch, 1988 ; Kintsch, & van Dijk, 1983 ; Rumelhart, 1980 ). For instance, Kintsche and van Dijk's (1983) situation model describes the comprehension process as arising from an interaction of three mental models: the reader's text representation, semantic or meaning-based representation, and situational representation (i.e., prior knowledge, experiences, and interest).

There is also empirical evidence for the importance of background knowledge in reading comprehension (e.g., Stahl, Hare, Sinatra, & Gregory, 1991 ). This may explain why children with SCD also have problems with elaborative inference making and comprehension monitoring ( Cain et al., 2001 ; Oakhill, 1984 , 1993 ; Oakhill & Yuill, 1996 ). Further, differences in the amount of background knowledge required across oral language and reading comprehension tasks may explain the present pattern of skill deficits. This explanation could be tested in future studies by having children perform reading comprehension and listening comprehension tasks on identical passages and have the tasks counterbalanced across two groups. However, deficits in background knowledge may not sufficiently explain why children have SCD. In some instances, children with SCD continue to perform below expectations even after background knowledge is controlled (e.g., Cain & Oakhill, 1999 ; Cain et al., 2001 ).

Regression to the mean

Another potential explanation for the discrepancy between the reading comprehension and oral language skills of children with SCD is regression to the mean. Across studies, children were selected on the basis of poor reading comprehension. This design can lead to an over-representation of children whose observed reading comprehension score is below their true score. Consequently, they will regress to their true score on almost any subsequent measure that is correlated with the original measure. In the present context, children who were selected on the basis of poor reading comprehension may perform less poorly on oral language due to regression to the mean. Future studies could test this hypothesis by administering a second reading comprehension measure and then comparing performance on this measure to oral language. Using another design that does not involve selection based on poor reading comprehension performance would also be helpful to rule out this explanation.

Theoretical and Practical Implications of the Findings

We began this article with a review of theories of reading comprehension. We now consider the implications of our results for the theories that we reviewed. We first consider our results within the simple view of reading framework. ( Gough & Tunmer, 1986 ; Hoover & Gough, 1990 ). Based on this framework, the view is that reading comprehension is the product of decoding and oral language comprehension. Our results are not consistent with the common version of the simple view in which reading comprehension is predicted by additive effects (i.e., main effects) of decoding and oral language comprehension. If the simple view is operationalized as the interaction (i.e., multiplicative effects) between decoding and oral language comprehension, however, the results could be considered consistent with this framework. Essentially, the oral language deficit of children with SCD interacts with their decoding to produce reading comprehension that is more impaired than would be accounted for by the simple main effects. This same logic would apply to interactive activation models of reading to the extent that the interactive activation is truly interactive.

As is emphasized by the simple view and interactive models of reading comprehension, oral language is a critical component of reading comprehension. This assertion is supported by the current findings and previous studies ( Kendeou et al., 2009 ; Roth et al., 2002 ). For instance, two studies included within the present meta-analysis, Catts et al. (2006) and Nation et al. (2004) , found that a substantial portion of children who are identified as having specific language impairment (SLI) also have coexisting reading comprehension difficulties. In both investigations, 30% or one-third of children with SCD were eligible for SLI identification. Even children who were not identified as having SLI were identified as having subclinical levels of poor language comprehension (Catts et al.). Children with SCD had very poor performance on the vocabulary measure and grammatical understanding task. Catts et al. and Nation et al. referred to this subclinical poor language comprehension as hidden language impairment because these children are not typically classified as having SLI. Yet, these impairments could still potentially lead to the comprehension problems observed in these children.

If we allow for the possibility of a latent decoding problem, then nearly all of the theories of reading comprehension could account for the pattern of results that were obtained. Similarly, if we allow for the possibility of differences between written and oral language, the results would be consistent with multiple theories of reading. It will be important to carry out research to determine the best explanation for the pattern of a greater deficit in reading comprehension than in oral language. The outcome of this research will potentially affect implications for theories of reading. For example, if the pattern of a greater deficit in reading comprehension than in oral language is found when (a) groups are matched on decoding fluency as well as accuracy, (b) the reading and oral language tasks are for equivalent material, and (c) the study design eliminates the possible confound of regression to the mean, the results would only be consistent with a theory of reading that had an interactive component in addition to whatever main effects might be represented.

The implications for practice are threefold. First, the results suggest that early oral language measures may serve as a means of identifying children who are at risk for later reading comprehension problems ( Cain & Oakhill, 2011 ; Justice et al., 2013 ; Kendeou et al., 2009 ; Nation & Snowling, 2004 ; Nation et al., 2010 ; Roth et al. 2002 ). Oral language weaknesses for children with SCD are evident fairly early on, are relatively stable over time, and are predictive of future reading comprehension performance (e.g., Cain & Oakhill, 2011 ; Justice et al., 2013 ; Nation et al., 2010 ). Thus, oral language measures can potentially serve as a screening method to identify which children have weaknesses in language skills. However, this must be approached cautiously because not all oral language measures are equally predictive of a child's future reading comprehension status. For instance, Tong et al. (2011) gave children with SCD morphological tasks that assessed derivational morphological awareness. Performance of readers with SCD in Grade 3 did not significantly differentiate children with SCD from those with normal reading comprehension in Grade 5. Yet, performance on this morphological task in Grade 5 did result in significant differences between the two groups. This suggests that measures of derivational morphological awareness, for instance, may not be ideal for assessing early oral language skills (see Nippold & Sun, 2008 ). Consequently, it is important to consider this when selecting potential screening measures.

Second, the findings suggest that children with deficits in critical oral language skills should receive targeted oral language instruction and intervention. Intervention studies focusing specifically on children with SCD have indicated that interventions containing an oral language component are more effective. For example, Clarke, Snowling, Truelove, and Hulme (2010) randomly assigned three groups of 8- and 9-year-olds with SCD to receive three different types of interventions: text comprehension training, oral language training (without reading or writing), and a combined text comprehension-oral language training format. All three groups showed reliable and statistically significant gains in reading comprehension compared to the control group; however, the group that received the oral language training maintained the greatest gains after an 11-month follow up (for a review, see Snowling & Hulme, 2012 ). These outcomes are also aligned with the findings of the present review. Thus, classroom instruction and intervention that incorporate elements that encourage comprehension proficiency, such as reading fluency ( NICHD, 2000 ) and oral language ( Snow et al., 1998 ), will likely be more effective at remediating reading comprehension difficulties.

Third, the current investigation highlights the need to develop a consistent operational definition of SCD (see Rønberg & Petersen, 2015 ). For studies included in the present investigation, there were multiple ways in which children with SCD were identified. Differences in identification criteria are potentially problematic because it can lead to over- or under-identification. Such differences can also potentially lead to different groups of children being identified as having SCD over time. Yet, variability in identification criteria is not exclusive to the present population of poor readers. There remains much discourse about this issue more broadly within the field of learning disabilities ( Mellard, Deshler, & Barth, 2004 ).

Limitations and Future Directions

There are several limitations of the present meta-analysis that must be addressed. First, the present review focused specifically on monolingual school-age children. Consequently, the results may not apply to second-language learner or adult populations. Second, several studies included in the present review used the Neale Analysis of Reading to assess reading comprehension and decoding ability without incorporating an additional measure of either skill. This is potentially problematic because both decoding and comprehension scores are obtained simultaneously as children read passages. Decoding problems could therefore affect comprehension scores (see Spooner et al., 2004 ). Third, we did not examine the effect of IQ on the obtained effect size estimates. It may be the case that variability in IQ may affect effect size outcomes. Fourth, it is important to acknowledge the potential presence of some publication bias for the between-group comparisons of reading comprehension and oral language. This may contribute to the larger deficits seen between these skills.

Another limitation of this meta-analysis is that it does not address possible causal relations between the deficits in oral language and reading comprehension. It is certainly possible that poor oral language skills may contribute to the deficits in reading comprehension; children must know a substantial portion of the words in a text in order to comprehend it ( Hu & Nation, 2000 ; Kendeou et al., 2009 ). However, it is also possible that poor reading comprehension constrains future vocabulary growth because text reading provides a basis for incidental word learning ( Cain et al., 2004 ). These relations may also be reciprocal (e.g., Wagner, Muse, & Tannenbaum, 2007 ). Additionally, the general absence of longitudinal data did not allow for a more comprehensive examination of the developmental delay versus deficit hypotheses. A final limitation of the present study is that it was limited to children who were monolingual speakers of their native language. It is increasingly common for children to know more than one language. Would the results of the present meta-analysis generalize to children who were second-language learners? We decided to answer this question by carrying out a similar meta-analysis of children with poor reading comprehension yet adequate decoding, but for children who were second-language learners ( Authors, 2017 ). Sixteen studies were identified that met inclusionary and exclusionary criteria. Hedge's g was used as the effect-size measure, random-effects models were used, and robust variance estimation was used to correct significance testing for dependent effect sizes. The results were remarkably consistent with those of the present meta-analysis. A deficit in oral language was replicated with an average weighted effect size of -0.80. The pattern of the deficit in oral language being only about a third as large as the deficit in reading comprehension was also replicated, with an average weighed effect size of -2.47. In summary, the pattern of results found in the present meta-analysis of studies whose participants were monolingual children generalize to children who are second language learners.

In conclusion, children who have SCD are typically impaired in oral language, but not to the degree they are impaired in reading comprehension. Consequently, the oral language impairment is not sufficient to explain the impairment in reading comprehension. Possible explanations for this pattern of results were considered, including a latent decoding deficit, differences between written and oral language, regression to the mean, and interactive effects. Testing these alternative explanations and others that might be considered represents a critical next step to advance our understanding of an important problem in reading.

Acknowledgments

This research was supported by Grant Numbers P50 HD52120 and 1F31HD087054-01 from the National Institute of Child Health and Human Development, Grant Numbers R305F100005 and R305F100027 from the Institute for Education Sciences, and a Predoctoral Interdisciplinary Training Grant Number R305B090021 from the Institute for Education Sciences.

Appendix A. 

Study descriptions and effect size estimates for children with specific reading comprehension deficits and typical readers (Study n = 86).

Note. RC = Reading comprehension; OL = Oral language; SCD = Children with specific reading comprehension deficits; TR = Typical readers.

Appendix B. 

Study descriptions and effect sizes for children with specific reading comprehension deficits compared with comprehension-age matched readers (Study n = 4).

Appendix C. 

Study descriptions and effect size estimates for within-child comparisons (Study n = 32).

Appendix D. 

Coding scheme for study type, participant age, and type of oral language measure.

Appendix E. 

Types of oral language skills assessed across studies (Study n = 86).

Note. For some studies, oral language was assessed but not explicitly reported.

2 For some comparisons, this comparison included skilled comprehenders.

3 Although groups were matched, correlations for the same measure between the two groups were not reported in most instances; thus, independent effect sizes were calculated.

4 Although groups were matched, correlations for the same measure between the two groups were not reported in most instances; thus, independent effect sizes were calculated.

5 In several instances, studies did not report correlations. For these studies, an estimated correlation was substituted.

6 We also conducted moderator analyses for type of oral language measure without accounting for hierarchical structure and the results remained the same [β #x0003D; −0.31, p > .05, t(1.40) = −0.98].

References marked with an asterisk indicate studies included in the meta-analysis. The in-test citations to studies selected for meta-analysis are not preceded by asterisks.

Factors and Problems Affecting Reading Comprehension of Undergraduate Students

This study aims to investigate factors affecting reading comprehension problems of 2nd, 3rd, and 4th-year students of English for International Communication (EIC) at Rajamagala University of Technology Lanna Tak. The study's objectives were 1) to examine the reading comprehension problems found most in 2nd, 3rd, and 4th year EIC students; and 2) to investigate the main factors influencing the reading problems that in turn, greatly affected the reading competence of 2nd, 3rd, and 4th-year students of EIC and how they cope with these problems. In this study, 77 EIC students demonstrated reading problems and factors which were adopted from Manutsawee (2015). The results showed that these students reflected different perceptions related to their reading problems and the factors that had an impact on their reading problems. The reading problems for the 2nd year EIC students were related to grammar, vocabulary, understanding, and personal experience with an average of 3.50, 3.43, 3.25, and 3.25, respectively. Meanwhile, the 3rd year students showed that they had difficulty with vocabulary (3.19) and grammar (3.10), with understanding and personal experience having the same average score (3.00). Finally, the 4th year students' reading problems were in the area of vocabulary (3.50), understanding and grammar (3.25), and personal experience (3.14). Moreover, the factor that affected EIC students' reading problems the most was identified as follows. The 2nd year students perceived students' attitude as the most influential factor at 3.91. However, the 3rd year students thought classroom teaching had the greatest impact on their reading problems at 3.79. Finally, the student's attitude was also the most influential factor, at 3.91, for the 4th year students.

• Related Documents

Investigating the Effective Factors of Using Personal Protective Equipment from the Perspective of Nurses Caring for COVID-19 Patients: A Cross-Sectional Study

Considering the importance of appropriate personal protective equipment (PPE) for preventing COVID-19 transmission, the aim of this study was to investigate the factors affecting the use of PPE from the perspective of the nurses caring for COVID-19 patients. This descriptive cross-sectional study surveyed 240 nurses working in the central COVID-19 hospitals of Arak, Iran. Nurses were enrolled in the study by a convenience sampling method. The data collection tool was a validated questionnaire. Data were analyzed by SPSS 16 software using descriptive statistics, analysis of variance (ANOVA), and independent sample t-test. Environmental (4.24 ± 0.45), personal (4.16 ± 0.42), and organizational (4.04 ± 0.50) factors all contribute significantly to nursing attitudes about PPE use (p < 0.05). The average score, combining all identified factors, was 4.15 ± 0.31. The most influential factor contributing to appropriate use of PPE was environmental, while the least impactful parameters were related to rules and regulations. Environmental factors have the greatest impact on the use of PPE from the perspective of the nurses caring for patients with COVID-19. Managers and healthcare organizations should provide appropriate and adequate PPE to nurses, educate them on proper use, and monitor the process to resolve barriers.

Exploring Undergraduate Students’ Views on Factors Influencing Academic Performance in Construction Related Disciplines in Bells University of Technology, Ota, Nigeria

Aims: There is usually dissimilarity in achievement as a result of several factors that influence the academic performance of students in universities. Therefore, this study aims to explore and investigate problem factors affecting the academic performance of undergraduate students in construction related disciplines in Bells University of Technology Ota, with a view to providing understanding on the major problem factors affecting their academic performance. Study Design:  Survey research design. Place and Duration of Study: Department of Building Technology, Bells University of Technology Ota, Ogun State, Nigeria, between May 2019 and January 2020. Methodology: The research employed a survey method with questionnaires distributed to undergraduate students in construction related disciplines in Bells University of Technology Ota. Data collected were analyzed using frequency, percentage, mean, rank and Kendall’s coefficient of concordance test. Results: Out of 172 questionnaires administered, 105 were sufficiently filled and returned, representing 61% of response rate. The results show that maturity, study strategies, interest in course, fear and stress and training and teaching style are the top five factors affecting academic performance of undergraduate students in construction related disciplines in Bells University of Technology Ota, Nigeria. Secondly, using SPSS (23), Kendall’s (Wa) value was found to be 0.036 at 0.000 significance level, therefore the study concluded that, there is statistically significant degree of agreement between different departments of the participants concerning their responses to factors that affect academic performance in construction related disciplines in Bells University of Technology, Ota, Nigeria. Conclusion: The study recommends that skipping of classes by students during early education should be discouraged so as to allow students attain maturity level for higher education. Secondly, universities should organize fear and stress management seminars and workshops for undergraduate students in addition to guidance and counseling sessions with a view to addressing psychological issues that may hinder their academic performance.

Role of Vocabulary Levels Test (VLT) in Chinese Undergraduate Students’ Comprehension of Textbooks

Vocabulary knowledge has always been considered crucial in reading comprehension. This paper aims to link Chinese learners’ performance on Vocabulary Levels Test (VLT) to the profiling of their textbook so as to draw pedagogical implications on English teaching. The participants were 108 freshmen in Jiangsu University of Technology, China. All of them took part in the VLT, and their English textbook was profiled to see the lexical level of the vocabulary. Results showed that 81.48% of the participants had mastered the 2,000-word level, and the vocabulary of the textbook was within the vocabulary knowledge of the participants. However, results also indicated that no participants had reached the mastery level of the K-3, K-5 and AWL words. Based on the findings of the research, it was suggested that the language teacher could make use of the textbook to consolidate the participants’ K-1 to K-2 words and at the same time give them more exposure to K-3 to K5 words as well as academic words by exposing them to some extra materials.

Students’ Speaking Skill in English Immersion School

This research aims to describe: (1) students' English speaking skill in English immersion school; (2) students’ strengths and weaknesses in speaking skill in English immersion school; (3) factors affecting students' speaking skill in English immersion school. The research was designed as a case study, conducted at a school applying English immersion in Bondowoso, East Java, Indonesia in Academic Year 2017-2018. The research used purposive sampling. Data were collecated by interview, questionnairre, observation, and document analysis. The informants were grade 2 students, English teachers, and the school founder. The analysis found that: (1) on the scale of 1-5 (1=very bad, 2=bad, 3=fair, 4=good, 5=very good), students’ English speaking skills were at levels of very good (2.22%), good (6.67%), fair (37.78%), bad (44.44%), and very bad (8.89%); (2) students’ strengths in English speaking skills were related to vocabulary with average score of 2.7, and the weaknesses dealt with fluency with average score of 2.25; (3) factors that were assumed to affect students’ English speaking skills were environment, local accent, vocabulary mastery, English use, preference, and habits. It is recommended that all parties involved in the English immersion program evaluate the running of the program and the content of the program to provide learning outputs through English immersion. Commitment to using English should be strengthened, especially in English Week program. Improvement on English teachers’ is needed by involving them in trainings. Finally, similar research at non-English immersion school is necessary to ensure the comparative results gained from both types of school.

Facilitating Successful Smart Campus Transitions: A Systems Thinking-SWOT Analysis Approach

An identification of strengths, weakness, opportunities, and threats (SWOT) factors remains imperative for enabling a successful Smart Campus transition. The absence of a structured approach for analyzing the relationships between these SWOT factors and the influence thereof on Smart Campus transitions negate effective implementation. This study leverages a systems thinking approach to bridge this gap. Data were collected through a stakeholder workshop within a University of Technology case study and analyzed using qualitative content analysis (QCA). This resulted in the establishment of SWOT factors affecting Smart Campus transitions. Systems thinking was utilized to analyze the relationships between these SWOT factors resulting in a causal loop diagram (CLD) highlighting extant interrelationships. A panel of experts drawn from the United Kingdom, New Zealand, and South Africa validated the relationships between the SWOT factors as elucidated in the CLD. Subsequently, a Smart Campus transition framework predicated on the CLD archetypes was developed. The framework provided a holistic approach to understanding the interrelationships between various SWOT factors influencing Smart Campus transitions. This framework remains a valuable tool for facilitating optimal strategic planning and management approaches by policy makers, academics, and implementers within the global Higher Education Institution (HEI) landscape for managing successful Smart Campus transition at the South African University of Technology (SAUoT) and beyond.

Job motivation and some factors affecting job motivation of office and support staffs at Dong Thap General hospital in 2017

The study was conducted with 02 objectives to describe and analyze some factors affecting the motivation of office and support staffs working at Dong Thap General Hospital in 2017. The research design is cross-sectional, combining quantitative and qualitative methods. Research results show that, the motivation average score of office and support staffs with the general motivation; burnout; job satisfaction; intrinsic job satisfaction; organizational commitment; conscientiousness; timeliness and attendance oscillate from 2.91 to 3.5. The motivation average score of office and support staffs with burnout is lowest and the motivation average score of office and support staffs with conscientiousness is highest. Some factors affecting the motivation of office and support staffs are income policies; training opportunities; satisfaction of colleagues and managers; working environment; equipment and facilities. The research results provide evidence to help the Board of Directors to develop solutions to improve the motivation of office and support staffs in the future. Keywords: job motivation, office and support staffs, hospital.

Factors Affecting Undergraduate Students’ Intention to Become a Chartered Accountant in Bangladesh

Purpose-The objective of this paper is to examine the factors affecting Bangladeshi students’ intention to be a chartered accountant. This paper specifically addresses factors like students’ gender, previous major at undergraduate level, CGPA, public vs private university status and family income.Design/methodology/approach - The study was conducted using a questionnaire survey with a sample of 271 students from five different public and private universities in Bangladesh. The 12x2"> test was applied to examine the relationship between these factors and students’ aspiration for a chartered accountant.Findings- The findings of the study show that out of five variables three variables, previous major, public vs private university status and family income, have significant relationship with students’ intention to pursue CA professional qualification. On the contrary, gender and CGPA of the students are considered as insignificant factors.Originality/Value-This research is the first study in Bangladesh which identifies several influencing factors for students’ choice to become a chartered accountant. The findings have implications for professional bodies in Bangladesh and other policy makers in their effort to develop strategies to motivate students to be a chartered accountant.

Tailoring Adsorption Induced Switchability of a Pillared Layer MOF by Crystal Size Engineering

The pillared layer framework DUT-8(Zn) (Zn₂(2,6-ndc)₂(dabco), 2,6-ndc = 2,6-naphthalenedicarboxylate, dabco = 1,4-diazabicyclo-[2.2.2]-octane, DUT = Dresden University of Technology) is a prototypical switchable MOF, showing characteristic adsorption and desorption induced open phase (<i>op</i>) to closed phase (<i>cp</i>) transformation associated with huge changes in cell volume. We demonstrate switchability strongly depends on a framework-specific critical particle size (d<i>_crit</i>). The solvent removal process (pore desolvation stress contracting the framework) significantly controls the <i>cp</i>/<i>op</i> ratio after desolvation and, subsequently, the adsorption induced switchability characteristics of the system. After desolvation, the dense <i>cp</i> phase of DUT-8(Zn) shows no adsorption-induced reopening and therefore is non-porous for N₂ at 77 K and CO₂ at 195 K. However, polar molecules with a higher adsorption enthalpy, such as the polar molecules such as chloromethane at 249 K and dichloromethane (DCM) at 298 K can reopen the macro-sized crystals upon adsorption. For macro-sized particles, the outer surface energy is negligible and only the type of metal (Zn, Co, Ni) controls the DCM-induced gate opening pressure. The framework stiffness increases from Zn to Ni as confirmed by DFT calculations, X-ray crystal structural analyses, and low frequency Raman spectroscopy. The partial disintegration of the Zn based node hinges produces an overall increased stabilization of<i> cp </i>vs. <i>op</i> phase shifts the critical particle size at which switchability starts to become suppressed to even lower values (d<i>_crit</i> < 200 nm) as compared to the Ni-based system (<i>d_crit</i> ≈ 500 nm). Hence, the three factors affecting switchability (energetics of the empty host, (<i>E_op-E_cp</i>) (I), particle size (II), and desolvation stress (III)) appear to be of the same order of magnitude and should be considered collectively, not individually.

The Contributions of Reading Strategies and Reading Frequencies toward Students’ Reading Comprehension Skill in Higher Education

This study was conducted because of the low level of students’ reading skill. The reading strategies and the reading frequencies were assumed as the factors affecting students’ reading skill. The aim of this study was to describe and to analyze the contribution of reading strategies and reading frequencies toward students’ reading comprehension skill. This study used the quantitative method with the descriptive correlational design. The population was 3000 students who attended Bahasa Indonesia subject in Universitas Negeri Padang, and 353 students were taken as samples by using proportional stratified random sampling. The instruments of this study were reading strategies scale, reading frequencies scale and a reading comprehension test which has been valid and reliable. Based on research findings and discussion, the following can be concluded: (1) reading strategies contributed significance and positively to the students' comprehension reading skill which is 19.6% (R = 0.443 R2 = 0.196 and p = 0.000), i.e. reading strategies contributed directly toward the skills of reading comprehension. Therefore, the higher the reading strategies, the better the reading comprehension skills of the students; (2) reading strategies contributed significance toward the students' reading comprehension skills of 33.1% (R = 0.575 R2 = 0.331 and p = 0.000), i.e. reading frequencies contributed directly to reading comprehension skills. Therefore, the better the reading frequencies, the better the reading comprehension skills of the students; (3) reading strategies and the reading frequencies together contributed significance toward the students' reading comprehension skills of 33.2% (R = 0.575, R2 = 0.332 and p = 0.000), i.e. reading strategies and reading frequencies together contributed directly to the comprehension reading skill. Therefore, the higher the reading strategies and reading frequencies will be the better the reading comprehension skill of the students.

Statistical Analysis of the Factors Affecting Academic Achievement of Undergraduate Students A Case Study of Faculty of Arts and Science Kufrah -Benghazi University

This study examined the factors affecting the academic achievement of undergraduate students at the faculty of Arts and Science Kufrah -Benghazi University as a case study. This study seeks to identify and analyze some determining factors that influence students' academic achievement in the study area. Four factors namely: psychological, family, learning facilities, and economic; were considered. The sample was randomly selected from the study population. A questionnaire was administered to 240 (90 males,150 females) students as a sample of this study. The responses of the students were analyzed to meet the objectives of the study using (SPSS) and displayed in forms and tables. After collecting the required data on the variables of the study, they were encoded to be entered into the computer to extract the statistical results. Statistical methods within the Statistical Package for Social Sciences (SPSS) were used to process data obtained by the field study of the sample. To analyze the data mean difference test is used. Results of arithmetic means of the psychological, family, learning facilities, and economic factors were medium. Furthermore, there were no statistically significant differences in the factors affecting academic achievement among the participants due to some demographic factors such as gender and marital status. following recommendations were made; provide proper learning facilities to the students and also improve the environment of the faculty. Furthermore, the students would perform well if they are properly guided by both their parents and teachers.

Export Citation Format

Share document.