Best Practice for ELLs: Screening

Level of evidence: Strong

This recommendation is based on a large number of studies that used reading assessment measures with English learners.

Brief summary of evidence to support this recommendation

Twenty-one studies demonstrated that three types of measures - phonological processing, letter and alphabetic knowledge, and reading of word lists or connected text - are valid means of determining which English learners are likely to benefit from typical classroom reading instruction and which children will require extra support. The primary purpose of these measures is to determine whether interventions are necessary to increase the rate of reading achievement.

These measures meet the standards of the American Psychological Association for valid screening instruments.¹

For students in kindergarten and grade 1. The early screening measures for kindergarten and the first grade fit into three categories:

• Measures of phonological awareness - such as segmenting the phonemes in a word, sound blending, and rhyming-are useful in both kindergarten and first grade.²
• Measures of familiarity with the alphabet and the alphabetic principle, especially measures of speed and accuracy in letter naming and phonological recoding, are useful in both kindergarten and first grade.³
• Measures of reading single words and knowledge of basic phonics rules are useful in first grade.⁴ Toward the middle and end of the first grade, and in the next few grades, measures of reading connected text accurately and fluently are useful.⁵

For students in grades 2 to 5. Three studies have demonstrated that oral reading fluency measures are valid screening measures for English learners and are positively associated with performance on comprehensive standardized reading tests. Oral reading fluency is emerging as a valid indicator of reading progress over time for English learners.⁶

These criterion-related validity studies are particularly important because another set of studies has investigated whether English learners can attain rates of reading growth comparable with those of their monolingual peers. These studies have demonstrated that English learners can learn to read in English at the same rate as their peers in the primary grades (K-2).⁷

Much of this evidence comes from research in Canada and from schools providing intensive and systematic instruction for all children, supplementary instruction for those falling behind, and instruction in settings where growth in oral proficiency is supported by both peer and teacher-student interactions. Evidence on reading interventions for English learners in the United States is the focus of Recommendation 2 (see Best Practice for ELLs: Small-Group Interventions).

How to carry out the recommendation

1. Districts should establish procedures for-and provide training for-schools to screen English learners for reading problems. The same measures and assessment approaches can be used with English learners and native English speakers.

Research shows that early reading measures, administered in English, can be used to screen English learners for reading problems. This finding is important because until recently it was widely believed that an absence of oral proficiency in English prevented English learners from learning to read in English,⁸ thus limiting the utility of early screening measures.

The common practice was to wait until English learners reached a reasonable level of oral English proficiency before assessing them on measures of beginning reading. In fact, oral language measures of syntax, listening comprehension, and oral vocabulary do not predict who is likely to struggle with learning to read.⁹ Yet research has consistently found that early reading measures administered in English are an excellent means for screening English learners, even those who know little English.¹⁰

It is very important to assess phonological processing, alphabet knowledge, phonics, and word reading skills. These measures, whether administered at the middle or end of kindergarten (or at the beginning of the first grade) have been shown to accurately predict later reading performance in all areas: word reading,¹¹ oral reading fluency,¹² and reading comprehension.¹³ So, it is essential to administer some type of screening to provide evidence-based beginning reading interventions to students in the primary grades.

In no way do these findings suggest that oral language proficiency and comprehension are unimportant in the early grades. These language abilities are critical for long-term success in school.¹⁴ We expand on this point in the importance of directly teaching academic English. The assessment findings point to effective ways to screen English learners for reading problems and to determine whether they are making sufficient progress in foundational areas of early reading.

2. Depending on resources, districts should consider collecting progress monitoring data more than three times a year for English learners at risk for reading problems. The severity of the problem should dictate how often progress is monitored - weekly or biweekly for students at high risk of reading problems.
3. Data from screening and progress monitoring assessments should be used to make decisions about the instructional support English learners need to learn to read.

Data from formative assessments should be used to modify (and intensify) the reading and English language development (or ESL) instruction a child receives. These interventions should be closely aligned with the core reading program. Possible interventions are described in Best Practice for ELLs: Small-Group Interventions.

Caveat: Measures administered at the beginning of kindergarten will tend to overidentify students as "at risk."¹⁵ A better indication of how students will respond to school instruction comes from performance scores from the middle and end of kindergarten. These scores should be used to identify students requiring serious instructional support. Scores from the beginning of kindergarten can provide a general sense of students' early literacy skills, but these scores should not be used as an indication of how well students are likely to respond to instruction.

4. Schools with performance benchmarks in reading in the early grades can use the same standards for English learners and for native English speakers to make adjustments in instruction when progress is insufficient. It is the opinion of the panel that schools should not consider below-grade-level performance in reading as "normal" or something that will resolve itself when oral language proficiency in English improves.

Using the same standards for successful reading performance with English learners and native English speakers may mean that a higher percentage of English learners will require more intensive reading instruction to reach the benchmarks, but we believe that this early emphasis on strong reading instruction will be helpful in the long run. Providing intensive early reading instruction for English learners does not imply they have a reading disability or they are not able to learn to read as well as other students. It means that while they are learning a new language and learning to read in that language simultaneously, they face challenges other students do not face. The instruction they receive should reflect the nature of this challenge.

A score on a screening measure indicating that an English learner may be at risk for reading difficulties does not mean the child has a reading disability. Being at risk means that the English learner needs extra instructional support to learn to read. This support might simply entail additional time on English letter names and letter sounds. In other cases additional support might entail intensive instruction in phonological awareness or reading fluency. Additional diagnostic assessments can be administered to determine what areas require instructional attention.

Unless districts have considerable resources and expertise, they should not try to develop the formative assessment materials on their own. Several screening and progress monitoring materials that have been developed and tested with native-English-speaking students are appropriate to use with English learners. Information about formative assessments can be found from a number of sources, including the Web and commercial developers. Please note that the authors of this guide did not conduct a comprehensive review of available assessments (such a large undertaking was beyond the scope of this project), and individual schools and districts should be careful when selecting assessments to use. It is important to select assessments that are reliable and valid.

5. Provide training on how teachers are to use formative assessment data to guide instruction.

The primary purpose of the formative assessment data is to determine which students are at risk (or not making sufficient progress) and to increase the intensity of reading instruction systematically for those students. We recommend that school-based teams of teachers be trained to examine formative assessment data to identify which English learners are at risk and to determine what instructional adjustments will increase reading progress. These teams can be for one grade or across grades. We believe that the reading coach, in schools that have one, should play a key role on these teams. Although principals should also play an important leadership role, it may be difficult for them to attend all meetings or be extensively involved.

Possible roadblocks and solutions

1. Some teachers believe that reading problems may resolve themselves once English learners develop proficiency in oral English. So, they are hesitant to refer these students for additional assistance or to provide intensive instruction in foundational areas of beginning reading.

There is no evidence to support the position that early reading problems experienced by English learners will resolve themselves once oral language skills in English are established.¹⁶ Districts should develop and disseminate materials explaining that using English oral language proficiency is as accurate as flipping a coin to decide which English learners are likely to have difficulty learning how to read.

To demonstrate that phonological, letter knowledge, and word reading measures are effective screening measures, principals and reading coaches can look at data from their own schools and see the links between scores on these measures in kindergarten and the first grade and later scores on state reading assessments.

2. Some teachers may feel that it is unfair to test a child in a language that she or he does not understand.

Although this is true in many areas, it is not true for tasks involving phonological processing, as long as the child understands the nature of the task.¹⁷ If students possess phonemic awareness of a word such as cake or fan, even without knowing the meaning they should be able to tell the examiner the first, middle, and last sounds in the word. Phonological awareness is an auditory skill that greatly helps students with reading development, and it transfers across languages. That is, if students learn the structure of sounds in one language, this knowledge will help them identify individual sounds in a second language without being taught explicitly what those individual sounds are. It is possible to demonstrate this to teachers by having them pull apart the sounds in words from an unfamiliar language, such as Russian or Arabic. Reading coaches can demonstrate that once a student knows how to identify the beginning, ending, or middle sound of a word, knowing the meaning of a word is irrelevant in being able to reproduce the sound.

Teachers should be clear that, for phonological processing tasks to be valid, English learners have to understand the task, but this is different from knowing word meanings. For an assessment to be valid the examiner must clearly explain the nature of the task and the child must understand what she or he is being asked to do. If possible, adults who are fluent in the child's native language can be hired and trained to administer assessments. But good training is essential. When appropriate, the examiner can explain or clarify the task in the language the child understands best. For districts with many native languages and few professional educators fluent in each native language, it is possible to make CDs of instruction in the appropriate native languages.

Make sure at least two or three practice items are provided before formal administration, when the task is modeled for the child and corrective feedback is provided. This will give all children (especially English learners) the opportunity to understand what the task requires of them. An important consideration for all assessments is to follow the testing guidelines and administration protocols provided with the assessment. It is acceptable to provide practice examples or explanations in the student's native language outside the testing situation. During the testing, however, it is essential that all assessment directions and protocols be followed. Remember, the purpose of the assessment is to determine whether children are phonologically aware or know the letters of the alphabet. It is not to determine how quickly or well children learn the formative assessment task when they are given explicit instruction in how to complete the task.

3. Some teachers may feel that native language assessments are more valid than English language measures for this group of students.

Formative early reading assessments in English are valid for English learners.¹⁸ If district and state policies permit testing a child in her or his native language, it is possible to get a richer picture of her decoding skills or familiarity with the alphabet. But this is not necessary for phonological awareness because it easily transfers across languages. Students who have this awareness in their native language will be able to demonstrate it on an English language assessment as long as they understand the task.¹⁹ In other words, even students who are limited in English will be able to demonstrate knowledge of phonological awareness and decoding in English.

4. Districts should anticipate that schools will have a tendency to view data collection as the terminal goal of conducting formative assessments, especially early in the process.

It is important to remind school personnel that data collection is just one step in the process. The goal of collecting formative assessment data is to identify students who are not making adequate progress and to increase the intensity of instruction for these students. In a system where the performance of all children is assessed multiple times a year, it is easy to become consumed by ways of organizing, analyzing, and presenting data and to lose sight of the primary purpose of data collection: to determine which students need extra support and which do not.

5. In districts that have the same early reading goals and standards for English learners and non-English learners, it is likely that the current performance of many English learners will be below these standards.

Although the average performance of English learners may be lower than that of non-English learners, there is no reason to assume that English learners cannot make the reading progress necessary to reach high standards of performance.²⁰ This progress will require providing more intensive instruction than the district might normally provide in both reading and language development.

6. Teachers may focus too much on what is tested-phonemic skills, decoding ability, and oral reading fluency-and neglect instruction in comprehension and vocabulary.

In monitoring student progress in phonological processing, phonics, and reading fluency, instruction in the development of comprehension and higher order thinking skills may be overlooked. But these skills should not be neglected. Instruction in comprehension and higher order skills should receive attention in the earliest phases of reading development. The challenge for schools will be to maintain a strong instructional focus on both higher and lower order skills.