Melby-Lervåg, M., & Lervåg, A. (2014). Reading comprehension and its underlying components in second-language learners: A meta-analysis of studies comparing first- and second-language learners. Psychological Bulletin, 140(2), 409-433. https://doi.org/10.1037/a0033890
Meta-analysis refers to a set of statistical techniques used to merge data from similar existing research studies to draw conclusions. The results of meta-analyses are held in high regard because they aggregate data from many independent studies and allow for statistical hypothesis testing, a method to determine the likelihood that a hypothesis is correct. The idea is that by summarizing many studies using rigorous techniques that take into consideration the probability of spurious findings, we can differentiate signal from noise, or meaningful and accurate information from random fluctuations in data. Meta-analysis can be used to address diverse types of research questions. In educational research, meta-analyses can help combine the results of many individual studies to evaluate the effectiveness of a particular intervention. For instance, the TCLD research team has conducted a series of meta-analyses examining the effects of reading interventions for students with or at risk for reading difficulties (e.g., Scammacca et al., 2007; 2015; Wanzek & Vaughn, 2007). Meta-analysis can also be used for other purposes. For instance, meta-analysis can be used to examine the relations between different variables, such as reading achievement and motivation as was done recently by Toste and colleagues (2020). Meta-analysis can also be applied to examine differences between groups of people. This was the case in Melby-Lervåg and Lervag’s study (2014), which serves as the focus of this summary.
Melby-Lervåg and Lervåg conducted a meta-analysis to compare reading performance differences between first- and second-language learners (FLL, or students reading in their native language, and SLL). In particular, the authors sought to understand differences between these subgroups in reading comprehension performance and in variables that contribute to reading comprehension ability, including language comprehension, decoding, and phonological awareness. This approach was grounded in a theory of reading called the simple view of reading. This theory helps to explain reading comprehension by breaking it into two broad domains—linguistic comprehension and decoding—with phonological awareness serving as a strong predictor of the latter. Language comprehension refers to the ability to understand meaning of language and is usually assessed in a listening task. Decoding involves the process of accurately and fluently pronouncing written words. Prior research suggests that second-language learners typically perform less well than first-language learners in reading comprehension, and that language comprehension plays a more outsized role for second-language learners than it does for first-language learners (Lesaux et al., 2006).
Understanding the similarities and differences between first- and second-language learners in terms of reading comprehension and its underlying skills provides important information about the strengths and challenges of second-language learners relative to primary language speakers. This may influence instructional decisions about the reading components most critical to address in order to reduce the achievement gap between primary and secondary language learners. According to the National Center for Education Statistics (2022), about one in five children in the United States speak a home language other than English, and one in ten are considered English learners due to a lack of proficiency in English. The prevalence of second-language learners and the significant differences in achievement found between first-and second-language speakers in reading comprehension make understanding the mechanisms behind the acquisition of reading comprehension a critical research priority.
Melby-Lervåg and Lervåg put forth three main hypotheses:
The authors carried out a systematic search process to identify articles that examined the reading comprehension, language comprehension, decoding, or phonological awareness of FLL and SLL. SLL were defined as (a) children who either use or study two language, and (b) are exposed to each language either regularly at home or in school. FLL were defined as children who spoke only one language at home, which was also the language of school instruction. The authors also coded for many potential moderators—variables that may explain the similarities and differences between English learners—including socioeconomic status, the language spoken at home, and several others.
The search yielded a total of 82 studies and 576 effect sizes that could be used in the meta-analysis. A majority of these studies were conducted in the U.S. with children who attended low-income schools and learned English as a second language. However, studies were also conducted in middle- and high-income schools in Canada, Europe, and Asia, studying both alphabetic and ideographic languages.
1. There would be a significant difference in reading comprehension performance between FLL and SLL; however, the size of this difference will vary based on children’s ages and the characteristics of the test used to assess reading comprehension. The authors posited that the differences between FLL and SLL will be smaller for young children because decoding is more important than language comprehension for young children. Their hypothesis related to the impact of test characteristics was based on prior research showing that tests of reading comprehension vary in the degree to which they are influenced by decoding and language comprehension (e.g., Francis et al., 2006).
1. FLL outperformed SLL in reading comprehension and this difference was moderate in size (g = .67). The difference between FLL and SLL was moderated by test type (passage comprehension tests led to greater differences than sentence comprehension and open-ended questions led to greater differences than multiple choice items. Language comprehension explained 30% of the variance in reading comprehension, and it played a larger role as students got older. Together, decoding a language comprehension explained 46% of the variance in comprehension.
2. Results would show large differences between FLL and SLL in language comprehension. They also predicted that these differences would be influenced by socioeconomic background, degree of exposure to the second language, and the linguistic overlap between the first and second language.
|2. Large differences were found in language comprehension (d = -1.31). Socioeconomic status, degree of second language exposure, and study origin influenced the differences between conditions. Greater differences between FLL and SLL were present for children from low-income backgrounds than children from middle- and high-income backgrounds. Children who spoke their first language exclusively at home also showed greater differences between condition than children who spoke both languages at home. Interestingly, the differences in language comprehension were smaller for studies conducted in Canada.|
|3. There would be small differences between FLL and SLL in phonological awareness and decoding that may favor the second-language learners, and that any group differences would be influenced by socioeconomic status, degree of exposure to second language, and whether the first language is alphabetic or ideographic.||3. No statistically significant difference between FLL and SLL were found in phonological awareness. A small yet statistically significant difference was found on decoding, favoring FLL. However, this pattern flipped when accounting for publication bias. An exception to this latter finding occurred in Canada, where second-language learners had greater decoding skills than first-language learners. This likely reflects differences between second-language learners in the U.S. and Canada and the schools that they attend that are not fully accounted for by socioeconomic status.|
There are at least a few important takeaways from this study. First, SLL perform lower in reading comprehension than FLL, on average. The primary cause of these difficulties is language comprehension. Differences in phonological awareness and decoding between SLL and FLL were not consistently found. The authors underscore the important of addressing language comprehension in order to address reading comprehension difficulties among second-language learners. The findings of the meta-analysis support this sentiment. It is worth noting, though, that although SLL do not appear to perform much differently than FLL on measures of phonological awareness and decoding, this does not preclude them from experiencing difficulties in these domains. Our research (Cho et al., 2019) with fourth-grade SLL with reading difficulties shows that these students demonstrate underperformance across language comprehension, decoding, and phonological awareness that contribute to their reading comprehension problems. This suggests that it would be wise to assess SLL decoding ability and provide an intervention when needed, even if we know, on average, most SLL will experience reading comprehension difficulties due to the developing second-language proficiency.
Second, it appears that the tests used to assess reading comprehension difficulties may influence how SLL perform relative to FLL. The differences between subgroups appears to be widest when comprehension tests require children to read longer passages and use open-ended response formats. This builds on prior research showing that it is critical to consider the type of reading comprehension test that is being used when interpreting findings (see the May 2017 Education Research Matters post on this topic), highlighting its importance when evaluating the performance of SLL. Third, the amount of second language spoken in the home appears to influence the size of reading comprehension difficulties between SLL and FLL. This suggests it may be critical to identify SLL whose families do not speak the second language in the home and monitor them more closely for reading comprehension difficulties.
Cho, E., Capin, P., Roberts, G., Roberts, G. J., & Vaughn, S. (2019). Examining sources and mechanisms of reading comprehension difficulties: Comparing English learners and non-English learners within the simple view of reading. Journal of Educational Psychology, 111(6), 982–1000. https://doi.org/10.1037/edu0000332
Francis, D. J., Rivera, M., Lesaux, N., Kieffer, M., & Rivera, H. (2006). Practical guidelines for the education of English language learners: Research-based recommendations for instruction and academic interventions. Center on Instruction. https://www.centeroninstruction.org/practical-guidelines-for-the-education-of-english-language-learners-research-based-recommendations-for-instruction-and-academic-interventions
Lesaux, N. K., Lipka, O., & Siegel, L. S. (2006). Investigating cognitive and linguistic abilities that influence the reading comprehension skills of children from diverse linguistic backgrounds. Reading and Writing, 19, 99–131. https://doi.org/10.1007/s11145-005-4713-6
National Center for Education Statistics. (2022). English Learners in Public Schools. Condition of Education. U.S. Department of Education, Institute of Education Sciences. https://nces.ed.gov/programs/coe/indicator/cgf
Scammacca, N., Roberts, G., Vaughn, S., Edmonds, M., Wexler, J., Reutebuch, C. K., & Torgesen, J. K. (2007). Interventions for adolescent struggling readers: A meta-analysis with implications for practice. Center on Instruction. https://www.centeroninstruction.org/interventions-for-adolescent-struggling-readers-a-meta-analysis-with-implications-for-practice
Scammacca, N. K., Roberts, G., Vaughn, S., & Stuebing, K. K. (2015). A meta-analysis of interventions for struggling readers in grades 4–12: 1980–2011. Journal of Learning Disabilities, 48(4), 369–390. https://doi.org/10.1177/0022219413504995
Toste, J. R., Didion, L., Peng, P., Filderman, M. J., & McClelland, A. M. (2020). A meta-analytic review of the relations between motivation and reading achievement for K–12 students. Review of Educational Research, 90(3), 420–456. https://doi.org/10.3102/0034654320919352
Wanzek, J., & Vaughn, S. (2007). Research-based implications from extensive early reading interventions. School Psychology Review, 36(4), 541–561. https://doi.org/10.1080/02796015.2007.12087917