Reading Wars, Reading Science, and English Learners

Abstract

Learning to read is more complex when learning a new language simultaneously.
Millions of English learners (ELs) in all-English programs face this challenge.
The author summarizes teaching ELs beginning and advanced reading skills in English.
The focus is on the similarities and differences between learning to read in a first and second language, with the latter focusing on limited English proficiency.
The goal is to integrate the science of reading to research on effective literacy instruction for ELs.
Promoting literacy requires using effective practices and helping ELs achieve higher English proficiency.
Both English literacy and oral language proficiency are crucial for ELs' success.
Learning to read is more complicated when done in a language one is learning to speak and understand.
The article examines teaching reading to ELs in English-only programs.
Teaching reading to native English speakers has been a source of controversy.
The science of reading is not the same as the science of teaching reading.
Hotly contested issues involve teaching writing and reading, extending beyond initial instruction.
Until recently, ELs have not been a focus in the reading wars.
Reasons include marginalization and focus on native-language literacy instruction.
Persistent poor achievement among ELs makes it urgent to have a clear view of literacy instruction in English.
Focus on ELs in English instruction doesn't ignore the benefits of native language literacy.
Native language instruction can positively affect English literacy and language achievement.
Bilingual education promotes bilingualism and biliteracy.
Focus on ELs in English instruction acknowledges that most ELs are in English-only programs.
These students face the dual challenge of learning literacy skills while learning English.
Exclusive focus on instruction doesn't ignore the challenges faced by ELs due to low socioeconomic status.
Challenges increase as ELs progress through school.
Some challenges can be addressed by improving instruction.
Other challenges require realignment of opportunities in society.
Socioeconomically rooted disparities are acknowledged, but research hasn't successfully addressed them.
Research has made progress with intensive interventions.
Stubborn gaps persist between ELs and non-ELs and between students from lower and higher socioeconomic strata.

The Science of Reading and Reading Instruction for ELs

The goal is to distill what is known about teaching ELs to read and develop as readers in English.
The focus includes ELs in early elementary grades learning to read for the first time.
Older ELs receiving literacy instruction in English are also included.
Target students are learning to read in a language they are still learning to speak and understand at a grade- or age-appropriate level.
The review does not include research where students are being taught to read in their native language.
Particular attention is paid to how English literacy instruction for ELs coincides with or diverges from the science of reading.
The intent is to build onto the knowledge base of the science of reading with research on effective literacy instruction for ELs.
The aim is to articulate a possible science of reading instruction for ELs.
An EL, as defined by Title III of the Elementary and Secondary Education Act of 1965 and amended by the Every Student Succeeds Act of 2015, is an individual who meets four criteria:
1. is aged 3 through 21;
2. is enrolled or preparing to enroll in an elementary or secondary school;
3. meets one of the following criteria—
  - was not born in the United States, or whose native language is a language other than English;
  - is a Native American or Alaska Native, or a native resident of the outlying areas; and comes from an environment where a language other than English has had a significant impact on the individual’s level of English language proficiency (ELP); or
  - is migratory, whose native language is a language other than English, and who comes from an environment where a language other than English is dominant—and
4. has difficulties in speaking, reading, writing, or understanding the English language, that may be sufficient to deny the individual
  - the ability to meet the challenging state academic standards;
  - the ability to successfully achieve in classrooms where the language of instruction is English; or
  - the opportunity to participate fully in society.

Reading Wars, Reading Science, and English Learners

Learning to read in a new language is similar to learning to read in a familiar language.
Differences exist, requiring extra support via English-language development (ELD) instruction.
Regardless of reading stage, effective literacy education for ELs is based on the science of reading and needs to be implemented as early as possible.

Learning to Read in a Second Language

Foundational skills and early reading are critical.
Learning to read in English requires understanding the alphabetic principle.
English has an opaque orthography, making it complex, but mostly regular.
Accurate, efficient word recognition is the bedrock of literacy development.
ELs must learn foundational skills without age- and grade-level command of English.
Successful early reading involves understanding words and texts.
Early stages emphasize accurate word recognition for fluent reading.
Neurolinguistic research shows similarities in learning to read in L1 and L2.
L2 reading involves systematic relationships between surface forms and meanings.
Brain activation networks are similar across languages in linking print and speech.
“Words and their meanings,” “speech processes,” and “syntactic binding” point to the challenges faced by ELs in learning to read in English.

What Beginning and Early Reading Supports Do ELs Need?

Two perspectives exist on necessary supports for ELs learning to read in a new language.
One perspective suggests minimal adjustments are needed if ELs understand directions and are taught some words explicitly.
The other perspective suggests ELs need more robust support, especially since it's harder to read if you don't know the language.
A third perspective suggests instruction should take cultural resources into account.

Institute of Education Sciences Practice Guide Recommendations

The first perspective was taken in an Institute of Education Sciences practice guide.
The guide made relatively little distinction between effective early reading instruction for ELs and non-ELs.
Most recommendations apply to both English speakers and ELs.
The practice guide emphasizes that assessments and standards are identical.
English-language proficiency is relatively unimportant for beginning and early literacy.
ELs experiencing early reading difficulties need the same instruction as non-ELs, just more of it.
The guide emphasizes interactive English-language development instruction for all ELs.
Supporting ELs’ English oral proficiency might affect literacy development at the beginning literacy stage.
Saunders, Foorman, and Carlson (2006) found that Spanish-speaking ELs in kindergarten classrooms with a block of time dedicated to ELD demonstrated greater growth in English oral language as compared with those in classrooms where teachers infused English oral language instruction throughout the day.
Saunders et al. also reported an unexpected finding: Students in classrooms with a dedicated English oral language development block demonstrated statistically significant higher English word identification scores on the Woodcock Language Proficiency Battery.
The effect, specifically on higher English word identification, was observed in English-only classrooms and in bilingual classrooms where students received Spanish literacy instruction.

Enhanced English Oral Language to Support ELs’ Early Literacy Instruction

Saunders et al.’s (2006) finding suggests that Gersten et al. (2007) might have underestimated the need for enhanced English-language learning opportunities to support ELs’ beginning and early English literacy development.
A closer look at the interventions reviewed by Gersten et al. provides further support for this possibility.
The randomized experiments reviewed by Gersten et al. (2007) involved three different interventions, all of which featured structured, systematic small-group direct instruction with emphases on phonological awareness, phonics, word recognition, use of decodable texts, and emphasis on fluency and comprehension.
Only a study by Vaughn et al. (2006) “demonstrated a statistically significant effect in reading” (Gersten et al., p. 24).
This interventionalso included the most extensive language support among those that Gersten et al. reviewed.
Vaughn et al. described two types of oral language support:
- Language support modifications: To ensure the students would understand and benefit from the Proactive Reading curriculum, we made language modifications to make concepts comprehensible.…
- Oracy and vocabulary development: Because participating students were bilingual, at risk for reading problems, and had low language proficiency scores in both English and Spanish…, we emphasized the development of oracy and vocabulary in English (the language of reading instruction).
The study demonstrated how a robust early literacy intervention that had previously been shown to be effective with non-ELs was made effective for ELs by adding English oral language instruction to help students understand the words and texts they were learning to read.
The fact that this intervention, with its greater emphasis on ELD in support of reading instruction, was the only one reviewed by Gersten et al. (2007) that obtained statistically significant effects on reading is very important and helps confirm what Saunders et al.’s (2006) findings suggested: Explicit oral ELD instruction helps support beginning English reading development for young ELs.
Precisely which of the oral language components in the intervention were responsible for helping ELs improve their early literacy skills is unknown, so future research could parse and study the individual components.

Similar Findings With Non-ELs

Similar findings with beginning and early readers who were not ELs were reported by Ehri, Dreyer, Flugman, and Gross (2007).
As with Vaughn et al.’s (2006) study, virtually all students came from low-income families and homes where Spanish was the L1.
However, students were not receiving EL services and probably had higher oral English skills than those in Vaughn et al.’s study.
As with Vaughn et al.’s intervention, Ehri et al.’s included explicit and systematic instruction of targeted knowledge and skills and daily lessons that included activities focused on the five core skills of phonological awareness, phonics, fluency, vocabulary, and comprehension.
The intervention also included significant attention to developing oral language and integrating oral language and vocabulary knowledge into the reading process: One purpose [of the intervention] was to develop oral language by encouraging students to talk about the books and by explaining the meanings of new vocabulary words.
These words were written in students’ personal books, and the meanings were reviewed each time the book was read.
Another goal was to spark the student’s interest in the book.
The tutor coached the student through the reading of the book and prompted the application of the reading skills and strategies that had been taught.
Students were encouraged to decode unknown words by relying on their letter–sound knowledge and then cross-checking with meaning and pictures to confirm the identities of the words.
Effects were moderate to large ( $0.44$ to $>1$ ) on measures of English phonological awareness, word reading, decoding, and reading and comprehending text.
Both interventions included principles of early reading instruction consistent with findings from research on the science of reading.
They both also incorporated oral English-language instruction designed to support students in learning to use and understand the English they were learning to read.
Results were comparable in the two studies.
Gersten et al. (2007) were undoubtedly correct that “intensive, interactive English language development instruction” (p. 2) is necessary for ELs.
Indeed, these English-language learning opportunities can and should start as early as preschool (e.g., Pollard-Durodola et al., 2018).
Yet, ELD instruction is important not just because “language abilities are critical for long-term success in school” (Gersten et al., 2007, p. 5).
ELD in support of early literacy is also important for beginning and early literacy success.
It is important for ELs and for language-minority students who, although not necessarily identified as ELs, are likely to benefit from an enhanced focus on oral English language (Ehri et al., 2007).
The English-language support that Vaughn et al. (2006) and Ehri et al. (2007) provided was not English- language instruction for general oral proficiency.
Rather, it was English oral language development to support the literacy instruction specifically.
Instructors defined words and clarified content using gestures or illustrations, clarified and modeled instructions (e.g., trace, copy), and read aloud texts, which students were reading, to emphasize listening comprehension and provide opportunities for oral responses using language in the texts.
ELs also need separate and additional comprehensive ELD instruction focusing on speaking and comprehending English more generally.
The oral language support required in the early stages of literacy acquisition is fairly narrow, given the words and texts being read.
As students move to more advanced stages of reading, they will require more advanced and complex oral English skills to support continuing and advanced literacy development.

The View From Neurolinguistics

Neurolinguistic research again helps in understanding Vaughn et al.’s (2006) and Ehri et al.’s (2007) findings.
Verhoeven, Perfetti, and Pugh (2019) suggested that L2 readers might “require additional supporting brain regions during learning” (p. 1).
According to Pugh (2004), students learning to read in their L1 “tend to rely on a large network of brain regions, as they try to bind orthography to an already-present knowledge of phonology and semantics” (p. 20).
The problem, of course, is that ELs have little or no “already-present knowledge of phonology and semantics” to which they can bind English orthography.
That is, the sound–symbol mapping system does not connect to already- present familiarity with speech sounds and known words of the language in which they are learning to read.
ELs have a great deal of linguistic (and general) knowledge in their L1, which can be used to bootstrap learning in English (Mallikarjun, Newman, & Novick, 2017).
Nonetheless, learning to read, which includes learning the idea of reading, is harder in a L2.
Li and Clariana (2019) provided a plausible explanation.
They reported that low L2-proficiency learners rely on more effortful cognitive processes to read and comprehend.
In contrast, higher L2-proficiency learners use their more developed and efficient L2 semantic network for comprehension.
As L2 learners become highly proficient in the reading language, they become more efficient readers and more similar to L1 readers: There is a shift from more effortful nonlinguistic processing to more automatic linguistic processing that “approximate[s] the brain network patterns of … L1 readers” (p. 103; see also Pugh, 2004).
These findings point to two conclusions:
- First, if ELs become more proficient in English, they become more efficient readers and more similar to native English speakers in their reading abilities.
- Second, to the extent that ELs do not become sufficiently proficient in English, their reading skills remain inefficient and their reading more effortful, which in turn are likely to have mutually reinforcing negative effects on achievement and motivation.

Beyond Foundational Skills and Beginning and Early Reading

Neurolinguistic and neurocognitive studies have shown that L2 learners’ early reading skills can develop nearly on pace with that of native speakers (Verhoeven, Voeten, & Vermeer, 2019).
The gaps between L2 learners and native speakers become more pronounced when language comprehension becomes more relevant for continued growth in reading beyond foundational skills (Raudszus, Segers, & Verhoeven, 2019).
Behavioral studies have converged on these findings.
A meta-analysis comparing phonological awareness, decoding, reading comprehension, and language comprehension and vocabulary of students learning to read in their L1 and L2 found that differences in phonological awareness and decoding were very small between L1 and L2 students (effect sizes of ~ $0.1$ ).
Differences in reading and language comprehension and vocabulary were much larger, with effect sizes of $0.6$ and $1.1$ , respectively (Melby-Lervåg & Lervåg, 2014).

The Simple View

Fluent Word Recognition and Language Comprehension
A foundational and widely accepted theory, the simple view of reading (Gough & Tunmer, 1986), indicates that reading proficiency is a function of two sets of skills:
1. accurate and fluent word recognition and
2. adequate language proficiency to comprehend texts.
According to this view, ELs will require greater emphasis on oral language development, particularly listening comprehension, as they progress through the grades so they can attain sufficient language proficiency, which in turn will permit reading comprehension (assuming accurate and fluent word recognition skills).
Moreover, accurate and fluent word recognition is partly dependent on knowing word meanings (Perfetti & Hart, 2002).
That is, if one knows the meanings of the words being read, one is more likely to read them accurately and fluently.
This then suggests oral language proficiency will play a role at all stages of reading development, relatively less at first, with word knowledge perhaps the most important, but increasing as students progress through the grades and English-language demands of reading and writing increase.
The theory remains equally applicable for both ELs and non-ELs.
The practical application differs because ELs will require more attention to developing oral proficiency than will students already proficient in English.
The school curriculum requires increasing language proficiency to keep pace with the increasing demands of textual and oral materials beginning around the middle elementary grades.
This is when ELs begin to show large and growing gaps as compared with English-proficient peers, particularly in language-intensive subjects.
Hence, the role of these factors is similar for all students, but the needs, and therefore what needs to be taught, grow at a rapidly increasing rate for ELs in the intermediate and later stages of reading development.

The Importance of Academic Language

Academic language (Bailey, 2007; Cummins, 1979) is a type of language critical for all students and particularly challenging for ELs.
In contrast to conversational or everyday language, academic language is more formal, abstract, and complex; assumes greater prior knowledge; and generally makes greater cognitive demands of both speakers and listeners (e.g., connectors such as moreover; discourse features such as generalizations followed by examples; abstract words such as dimension and concept; discipline-specific language such as technical terms found in a lab report or analytical essay, such as asymptote, zoonotic, plutocracy, topology, and metaphor).
Measures of ELs’ academic English, such as the Woodcock Language Proficiency Battery, have a much stronger association with reading achievement than measures of everyday language, such as the Basic Inventory of Natural Language (Genesee, Lindholm-Leary, Saunders, & Christian, 2006; for more on academic English and its role in ELs’ literacy development as they progress through the grades, see Goldenberg & Coleman, 2010).
One of the reasons academic language is so important is its relation to knowledge, another factor in the reading comprehension process (Perfetti & Stafura, 2014).
Academic language facilitates the acquisition and use of disciplinary and general knowledge required for success in school (e.g., Bailey, 2007; Janzen, 2008; Snow & Uccelli, 2009).
As Janzen (2008) observed, “the language of oral interaction does not necessarily have to be English” (p. 1030).
Yet, eventually, when students are in an English academic context, they must have adequate command of academic language in English.
This will be true even if they have the benefit of long-term bilingual education or, even more rarely, are in an environment that promotes simultaneous use of more than one language (i.e., translanguaging; Creese & Blackledge, 2015).
The evidence documenting the relation between knowledge, whether conceptualized as domain, background, or general, and reading comprehension is voluminous.
Researchers have even been able to establish a knowledge threshold, such that knowledge facilitates reading comprehension only when knowledge reaches a certain threshold (O’Reilly, Wang, & Sabatini, 2019).
Below it, readers do not have enough knowledge to move the comprehension needle appreciably.
As students progress through school, knowledge becomes an increasingly important factor in reading comprehension.
A little knowledge, and even a bit more than a little, is usually insufficient for comprehending challenging and complex texts (O’Reilly et al., 2019).
Some threshold of knowledge must be reached in order for students to begin fully comprehending texts they start encountering in upper elementary school.
Above the threshold, which likely varies by topic, type of knowledge indexed, text level, and other variables, more knowledge leads to greater comprehension.
Below the threshold, “a lack of knowledge becomes a barrier to comprehension” (O’Reilly et al., 2019, p. 1344).

EL Supports Beyond Beginning and Early Reading

As was true for beginning and early readers, studies of older ELs have also suggested similarities between effective practices for ELs and non-ELs.
Cheung and Slavin (2005) reviewed many of the same programs, through grade 5, as the National Literacy Panel on Language- Minority Children and Youth (August & Shanahan, 2006) and concluded that effective programs for ELs were “typically, programs originally designed for use with English- proficient students…[and] considerably adapted for use with ELLs, with more emphasis on vocabulary and oral language (see Fitzgerald, 1995; Slavin & Calderón, 2001)” (p. 263).

Balanced Literacy

A Cautionary Tale
O’Day’s (2009) observational study of third- to fifth- grade students in a balanced literacy program provides a cautionary tale about what can happen for ELs when classroom instruction has a “pervasive emphasis on meaning-making and comprehension…[and] an emphasis on meaningful interaction with and communication about text” (p. 99), but it does not take into account the English proficiency needs of ELs.
Two sets of findings are important.
The first is that three variables—focus on higher level text meaning, writing composition and discussion, and accountable talk—statistically significantly predicted reading comprehension growth for non-ELs.
For ELs, the correlations were positive but not statistically significant.
O’Day suggested this explanation: The balanced literacy program was assumed to “promote oral language fluency, as well as conversational and academic vocabulary development,…[which then] should provide an additional indirect benefit to raising EL students’ reading comprehension” (p. 99).
Yet, instead of promoting ELs’ ELD and providing “an additional indirect benefit” of improved reading comprehension, O’Day (2009) suggested that ELs “did not have the foundational English language skills to take advantage of the opportunities for meaning construction to the same extent that native speakers did” (p. 105).
She hypothesized “that at least some of the questioning, writing, and accountable talk discussion in San Diego classrooms was occurring at a linguistic level that was simply beyond the ability of many of the EL students to comprehend” (p. 105).
The program was presumed to provide the benefit of improving ELs’ English-language skills, yet not only did it not provide this benefit; ELs’ lack of English proficiency actually kept them from benefiting from the instruction.
The direction of effects was the same as that for non-ELs, but the strength of the association was weaker to the point of statistical nonsignificance.
O’Day’s (2009) second important finding was that a fourth variable—“providing students with explicit information through telling” (p. 106)— predicted achievement for both groups significantly but in opposite directions: positive for ELs and negative for non-ELs.
This disordinal interaction suggests that too much teacher talk might be counterproductive for English-proficient students, but in contrast, ELs seem to benefit from more explicit instruction and information.
This finding echoes Kucer and Silva’s (1999), who examined the efficacy of inductive teaching for language-minority students from low-income families.
Kucer and Silva found that students in a whole- language classroom made literally no progress in some areas of literacy learning, such as writing and writing conventions, and suggested that more explicit instruction, or telling, might have been more productive.
This, of course, does not mean that ELs always or only require explicit teaching, nor that they are incapable of inferencing and critical thinking.
What it means that teachers providing instruction in English must be careful that the language be comprehensible to ELs.
O’Day’s (2009) results also suggest that claims about benefits to ELs’ oral language proficiency made by some balanced literacy programs should be treated cautiously (see also Student Achievement Partners, 2020).
Instruction that is effective for non-ELs can be effective for ELs as well, as we have seen, but it needs to take into account ELs’ language needs.
O’Day’s study suggested two important adaptations: explicit instruction when needed (i.e., providing information through telling) and assuring that questioning, writing, and classroom discussion be carried out at a linguistic level that is accessible to ELs, which of course should include directly helping ELs acquire the necessary language.

English-Language Supports for ELs

In contrast to the lack of English-language support O’Day (2009) found in the balanced literacy program, several interventions undertaken with ELs in grades 5–7 have demonstrated that a sound instructional program, augmented with an ELD component designed to help ELs learn the academic content, helps ELs develop language and literacy skills in intermediate stages of literacy development (August, Branum-Martin, Cardenas-Hagan, & Francis, 2009; Carlo et al., 2004; Lesaux, Kieffer, Faller, & Kelley, 2010; Vaughn et al., 2009).
Results were relatively modest and, surprisingly, nearly identical for ELs and non-ELs.
The interventions have in common well-structured, planned, and delivered curriculum; intense, ongoing teacher training; and explicit instruction presenting conceptually challenging grade-level material and corresponding academic vocabulary.
The interventions used a number of strategies, sometimes referred to as sheltered instruction (Echevarría, Vogt, & Short, 2016) or integrated ELD (Ybarra & Hollingsworth, 2017), such as videos, graphic organizers, paired and group work, and interactive scaffolded discussions.
Instruction was essentially all in English.
The interventions by August et al. (2009) and Vaughn et al. (2009) also used L1 support such as Spanish cognates (e.g., infirm/enfermo) or quick translations to promote comprehension and access to content for Spanish-speaking ELs.
These are all recommendations suggested by a second Institute of Education Sciences practice guide (Baker et al., 2014).
Aside from Carlo et al. (2004), none of the interventions described culturally relevant instructional components other than L1 support.
Carlo et al. (2004) is the earliest of this group of studies targeting academic language, specifically, academic vocabulary, to improve students’ reading comprehension.
It is an important and widely cited study that looked at the effect of a vocabulary intervention on both ELs and non-ELs.
The intervention comprised generally acknowledged sound instructional practices combined with ones considered culturally and linguistically appropriate for ELs.
Units were constructed around the topic of immigration (students and/or parents were Latin American or Caribbean immigrants), with readings from newspaper articles, diaries, firsthand documentation of the immigrant experience, and historical and fictional accounts and audio and written Spanish versions of the texts.
The program showed modest positive effects on most of the vocabulary measures, although “the impact on comprehension was much lower than that on word learning” (August & Shanahan, 2006, p. 431).
There were some effects at some sites, and it was not possible to gauge overall effects across sites (August & Shanahan, 2006, Table 15.1).
The impact on ELs’ reading comprehension was modest but still supported the hypothesis that enhanced English-language support, specifically, academic vocabulary, helps support reading development in late elementary school.
It is important to note that the program included far more than academic-language support and culturally relevant content.
There were whole-group lessons using vocabulary instruction as recommended by Beck, McKeown, and Kucan (2002), such as depth-of-word activities and study of words in context, heterogeneous language groups, and systematic reviews of words taught.
Teacher materials included detailed lesson plans with quasi-scripted lesson guides and ancillary materials, such as overheads, worksheets, homework assignments, and weekly review tests.
Teachers and researchers held biweekly meetings meant to provide support to the teachers throughout the implementation of the curriculum.
The program’s overall effects were positive but could have been the result of any combination of features.

EL Supports Worked for Non-ELs, Too

and Do Not Address the Achievement Gap
It is also important to note that the impact of all four programs was nearly identical for ELs and non-ELs.
Consequently, because ELs scored lower before the intervention, they also scored lower afterward, “reflecting the generally higher performance of native English speakers” (Carlo et al., 2004, p. 196).
The intervention did not close the gap, as the article’s title suggests, unless we compare treatment ELs with control non-ELs, a finding reported by Schechter, Macaruso, Kazakoff, and Brooke (2015) in a study of blended learning: “At pretest, ELL treatment students scored statistically significantly below non-ELL control students…. By posttest, ELL treatment students closed the performance gap and showed nearly identical scores to non-ELL control students” (p 193).
Even with improvement from the interventions, sometimes ELs did not catch up with non-ELs.
August et al. (2009) found that at posttest, non-ELs in the control condition had higher scores in vocabulary than did ELs in the treatment condition.
In a subsequent study also focusing on academic language to promote literacy in science, August et al. (2014) found weaker effects for ELs as compared with controls (effect size= $0.16$ ) and much greater impact of the intervention for non-ELs as compared with ELs (effect size = $0.80$ , in favor of non-ELs, as calculated by Estrella, Au, Jaeggi, & Collins, 2018; see below for more on this meta-analysis).
In short, the good news is that ELs in these studies benefited from the interventions designed to bolster their academic language, specifically, vocabulary in the content areas, which in turn produced at least modest positive effects on reading comprehension.
The more cautious news is that the interventions did nothing to address the achievement gap between ELs and non-ELs, which has been a driving motivation in EL research, practice, and policy for over 50 years.
The gaps disappeared only if the interventions were withheld from non-ELs.
A possibly even more troubling finding was reported in a meta-analysis of inquiry science education for ELs and non-ELs, in which August et al.’s (2009, 2014) studies were included.
Estrella et al. (2018) found that inquiry science instruction for ELs was modestly more effective in promoting learning of science knowledge and literacy as compared with traditional or business-as-usual instruction (mean effect size = $0.28$ ).
The effect increased to $0.32$ when professional development (PD) for teachers addressed instructional needs of ELs.
Absent EL-oriented PD, the effect size for ELs dropped to a statistically non-significant $0.06$ .
Non-ELs benefited more from science inquiry than ELs did by the same order of magnitude ( $0.31$ ) as ELs benefited from inquiry science instruction as compared with ELs in control conditions (Estrella et al., 2018).
The meta-analysis produced some important and startling findings about the role of EL-oriented PD.
First, EL-oriented PD fully explained the positive effects of inquiry science for ELs: Without EL-oriented PD, inquiry science showed no advantage over traditional or business-as-usual instruction for ELs.
Second, when there was EL-oriented PD, non-ELs benefited more than ELs; the effect size comparing non-EL and EL inquiry science achievement was $0.45$ in favor of non-ELs.
Without EL-oriented PD, the effect size was only $0.19$ .
The stronger effect of EL-oriented PD for non-ELs, added to the findings of equivalent effects of language and literacy interventions for ELs and non-ELs, raises questions about whether there is anything unique to ELs that requires specialized EL-focused PD and, by extension, instruction.

Limitations

The principal limitations to the available research base are, first, that classroom evidence comes almost exclusively from studies with Spanish-speaking ELs from low- income families and, second, that most of the studies were undertaken with grades K–8.
Relevant high school studies are very rare.
If we are to work toward something like a science of reading instruction for ELs, we must expand the number and ages of populations studied.

Findings, Implications, Questions, and Priorities

With the preceding as caveat, here are some things the field can claim based on the available data:
- First, learning to read in a language with an alphabetic orthography such as English is very similar for English speakers and for ELs.
- Both require skills and understandings about how speech and print are connected through the alphabetic principle, how letters represent speech sounds, and how written words represent spoken words.
- Second, because ELs are simultaneously learning to read in English while learning to speak and understand it, they need additional instruction and support in learning the English vocabulary they are expected to learn to read.
- For beginning readers, word meanings provide support, not to read the words but to recognize and confirm when they are read (decoded).
- Oral language instruction, so that words are comprehensible, is helpful, probably even necessary, for beginning EL readers but not for students already proficient in English.
- Third, as students progress through the grades, they are expected to read and comprehend increasingly complex and demanding texts.
- The literacy-learning agenda must accordingly become more complex and demanding for all students.
- For ELs, the English oral language instruction and support must also become more complex and demanding.
- Academic language, in particular, becomes increasingly important for literacy development because students must understand the language of texts and acquire the general and disciplinary knowledge required for text comprehension.

Implications for the Science of Teaching Reading for ELs

The implication for the science of reading, or more precisely, a possible science of teaching reading for ELs, is quite clear: What is known about effective literacy instruction for non-ELs is the foundation of effective literacy instruction for ELs.
However, attention must be paid to students’ English oral language proficiency as it relates to what students are expected