Good Instruction is Good for Everyone—Or Is It? English Language Learners in a Balanced Literacy Approach

Abstract

Prior research suggests that many of the elements of effective instruction for native speakers are beneficial for English learners as well. But are specific strategies that appear effective for native speakers equally so for nonnative speakers? Are other companion efforts necessary? This article explores these questions by examining the comparative effects of selected literacy instructional practices on reading comprehension for English learners and non-English learners, using data from a 3-year study of the implementation and effects of a balanced literacy approach in San Diego City Schools. Then, using qualitative data gathered over 2 years from 133 teachers, as well as school administrators and instructional coaches in 9 case study elementary schools, the article considers what might have been missing in the district's approach to its over 24,000 elementary English learner students and how the strengths that were developed during the district's literacy reforms might be used to address continuing weaknesses.

ACKNOWLEDGMENTS

The research for this article was supported through generous grants from the William and Flora Hewlett Foundation, the Bill and Melinda Gates Foundation, and the Atlantic Philanthropies. I extend my gratitude to the funders, the district and school staff in SDCS, our data collection and analysis team (particularly Catherine Bitter, Miguel Socias, Paul Gubbins, and Heather Quick), and to Libi Gil and Marshall Smith, who provided comments and feedback on prior versions of this article. However, I bear sole responsibility for the findings and interpretations presented here.

Notes

¹Nomenclature for students in U.S. schools who are dominant in a language other than English varies across time and jurisdiction. At present, the most popular terminology nationally is English language learner or ELL, and NCLB refers to this group of students as Limited English proficient (LEP). Since the passage of Proposition 227 in 1998, the official designation of such students in California is English Learner (EL), which refers to students whose primary language is not English and who have not yet achieved fluent English proficiency (as measured by the CELDT) and/or reached a threshold level on the California Standards Test (CST) in English Language Arts or other criteria established by local district. For consistency, I have adopted the California nomenclature throughout this article to refer not only to students in San Diego and the rest of the state who have been designated as ELs, but also comparable students nationally.

²Analyses of a national school AYP database reveal that students with disabilities and students with LEP were the subgroups most likely to miss AYP targets. Thirty-seven percent of schools held accountable for the students-with-disability subgroup and 26% of those accountable for LEP students missed their targets for those groups in 2003–2004 (LeFloch et al., 2007). Those numbers are expected to increase as the required AYP targets move toward 100% proficiency in the coming years.

³In 2001–2002, 60% of EL students nationally were receiving essentially all of their instruction in English (Goldenberg, 2008). Given recent trends in several states, including California, this percentage is likely higher now.

⁴In California, the reclassification process is determined locally, using four criteria established by the State Board of Education: assessment of English language proficiency on the California English Language Development Test (CELDT), teacher evaluation of a student's academic performance, parent opinion and consultation, and performance in basic skills as measured by the CST in English language arts. The specific measures and cutoffs for these criteria, as well as the process for reclassification, may vary from district to district and from one time point to another, thus affecting which students are considered to be in the EL subgroup. For example, between 2003 and 2004 Long Beach Unified made a major push to reclassify more students from EL status to that of R-FEP. The percentage of EL students reclassified jumped from 6.4% to 16.9% in that one year and then continued to rise the following year. In 2005, Long Beach Unified's reclassification rate was twice that of the state as a whole. As higher performing students are reclassified out of the EL subgroup, which is also continually replenished with new immigrants (who are likely to have lower test scores), cross-sectional analyses of the scores for the EL subgroup may appear to decline when, in fact, longitudinal student level analyses would show considerable progress. To the extent that policies and rates for reclassification differ from district to district, resulting percentages of proficient EL students will not be comparable.

⁵We considered various explanations for the observed discrepancies in the findings for the first and second year of the study, including the timing of the visits with respect to test preparation activities, the separation of writing instruction from other literacy activities in some schools during the second year, and changes in characteristics of the student population. The most promising of these was the increase in ELs in the relevant grades in Year 2.

⁶In 2004–2005, we visited each classroom in fall, winter, and spring; however, in 2005–2006, we only visited classrooms twice, in fall and spring. The analyses reported here are for 2004–2005, as we determined that these data were more robust and fully reflective of the instruction taking place over the course of the year.

Notes. ∗p < 0.10, ∗∗p < 0.05, ∗∗∗p < 0.01. DRP = Degrees of Reading Power; NCE = Normal Curve Equivalent; ELA = English Language Arts; CST = California Standards Test.

^a DRP NCE 2004–2005 is the Normal Curve Equivalent Score for students in the spring of 2004–2005.

^b English learners are compared against students who are classified as Redesignated Fluent English Proficient, Initially Fluent English Proficient, and English Only.

^c CST ELA 2003–2004 is the CST total scale score in ELA for students in the spring of 2003–2004. This score has been standardized within grade for students in the sample.

^d Students identified as White, Black, and Other Ethnicity are compared against Hispanic Students. Other Ethnicity includes Asian, Pacific Islander, and American Indian students.

^e Measures the percentage of 5-min segments in which the given literacy activity was coded. This percentage has been standardized within the pool of teachers in the sample.

^f Accountable Talk is an index calculated by averaging all eight measures of accountable talk across segments and observations. This percentage has been standardized within the pool of teachers in the sample.

Notes. ∗p < 0.10, ∗∗p < 0.05, ∗∗∗p < 0.01. EL = English Learner; DRP = Degrees of Reading Power; NCE = Normal Curve Equivalent; ELA = English Language Arts; CST = California Standards Test.

^a DRP NCE 04–05 is the score for students in the spring of 2004–2005.

^b English Learners are compared against students who are classified as Redesignated Fluent English Proficient, Initially Fluent English Proficient, and English Only.

^c CST ELA 2003–2004 is the CST total scale score in ELA for students in the spring of 2003–2004. This score has been standardized within grade for students in the sample.

^d Students identified as White, Black, and Other Ethnicity are compared against Hispanic Students. Other Ethnicity includes Asian, Pacific Islander, and American Indian students.

^e Measures the percentage of 5-min segments in which the given literacy activity was coded. This percentage has been standardized within the pool of teachers in the sample.

^f Measures the percentage of 5-min segments in which the given interaction pattern for the teacher was coded. This percentage has been standardized within the pool of teachers in the sample.

⁷In this model, the accountable talk code has been replaced with the code for discussion/conversation. These two variables were correlated and the effects of both were washed out when the two were analyzed together. The main distinction (as discussed later) is that the discussion/conversation variable did not include any inferences about the focus on ideas or the press for evidence.

⁸Proposition 227, passed in 1998, the same year that Alan Bersin and Anthony Alvarado took the helm in San Diego, required that

English learners entering California Schools be placed in structured English immersion for a period ‘not normally to exceed one year,’ then be transferred to mainstream classrooms taught ‘overwhelmingly in English’…The law also included a provision allowing parents of English learners to opt for bilingual education via waivers. (Parrish et al., 2006, pp. 3–4)

⁹Alan Bersin's superintendency ended in June 2005 and the district is now on its fourth superintendent (including two interims) since that time. Our data indicate that, in 2005–2006, the balanced literacy reforms were still in force in the elementary schools in the district—at least those serving much of the EL student population.