Four-Year High School Graduation Rate of Students in Arizona by English Learner Status

• In Event: Beyond the Homogeneous English Learner Classification: New Progress Analyses Using New, Multiple English Learner Subgroups

Abstract

Objectives
The study examines: 1) the four-year high school graduation rates for five student subgroups by their English language learner (ELL) status at the start of 9th grade; and 2) how the graduation rates differed across the subgroups with and without controlling for student demographic characteristics and prior academic achievement. These subgroups include: long-term ELLs, new ELLs, recently proficient former ELLs, long-term proficient former ELLs, and never ELLs.

Theoretical framework
ELLs as a group tend to lag behind native English speakers in academic outcomes, including high school graduation (Kindler, 2002; Olsen, 2010; Ruiz-de-Velasco & Fix, 2000; Short & Fitzsimmons, 2007; Venezia et al., 2005). The difficulty of simultaneously learning English and content knowledge appears to be more acute for ELLs at the secondary level (Cook, Wilmes, Boals, & Santos, 2008; Grissom, 2004; Kieffer, 2008, 2010, 2011; Salazar, 2007).
Studies suggest that current ELLs and former ELLs at the secondary level should be further broken down to more differentiated subgroups when examining their academic performance: long-term ELLs and new ELLs for current ELLs; recently proficient ELLs and long-term proficient ELLs for former ELLs (e.g., Gwynne et al., 2012). Knowing more about the variation in achievement among these groups of students, especially at the secondary level, will enable educators to better support ELLs and former ELLs to graduate on time.
Data source
The analytic data come from Arizona Department of Education with a cohort sample size of 63,144 students in 9th grade in 2010/11 followed through 2013/14 (expected grade 12/graduation).

Methods
The observed four-year high school graduation rates are calculated using descriptive analysis. In addition, the graduation rates are estimated from a two-level logit model with possibility of on-time high school graduation as the outcome and student subgroup membership as the main effect, controlling for student demographic characteristics and prior academic performance.

Results
The study found that the four-year high school graduation in Arizona varied across the student subgroups. Before controlling for student characteristics, never ELLs had the highest graduation rate. Yet, after controlling for certain student characteristics, ELLs who had been reclassified in elementary school had the highest graduation rate. Prior academic achievement may have been a key factor. The graduation rate among ELLs with demographic characteristics similar to the Arizona average was no more than 55 percent. On the other hand, the graduation rate among ELLs with both demographic characteristics and prior academic achievement equal to the Arizona average exceeded 80 percent. Regardless of their background characteristics, the earlier ELLs achieved English proficiency, the higher their graduation rates. ELLs who were not reclassified before high school had the most difficulty in graduating. (See Appendix B).

Scientific significance of the study
No studies have examined the high school graduation rate across ELL status subgroups based on state-wide data. This study adds to the research literature by providing empirical evidence on the relationship between on-time high school graduation and students’ membership in different ELL status groups at their entry into 9th grade in Arizona.

Authors

• Min Huang, WestEd

• Eric Haas, WestEd

• Niufeng Zhu, WestEd