Using the ACCESS test as a predictive measure of English learner success on the biology end-of-course-test in Georgia

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
University of Alabama Libraries

In order to be successful in an educational setting, a student must acquire a particular register of language specific to academia, or an academic register. It cannot be acquired through ordinary social intercourse. It is specific to school. Acquiring the academic register is the greatest challenge English Learners (EL) face during their academic careers. The federal government has recognized that challenge and has placed the burden, where it rightly belongs, on educators. With the passage of the No Child Left Behind Act of 2001, schools, districts, and states are accountable for the annual progress of ELs' language proficiency with an emphasis on the acquisition of the academic register. The goal is for all students to attain a high school diploma. In order for ELs to do this, they must not only learn vocabulary related content and concepts, but the language that crosses content areas such as analyze,' infer,' and summarize.' They must learn multiple meanings of seemingly simple words, such as like,' left,' and right,' which cause confusion for those who only know the more common definition in their social register. English Learners are required to take federal and state mandated assessments of their content knowledge and their English language proficiency. In an era of data-informed instruction, these sets of data are available to the educator in order to best inform decisions related to the individual needs of each student. This study sought to identify a correlation between academic language proficiency as measured by the Assessing Comprehension and Communication in English State-to-State for English Language Learners (ACCESS) test, 2007 edition, and the Biology End-of-Course-Tests (EOCT) for 164 ninth graders at one high school in Georgia between the years 2007 and 2012. A strong, positive correlation was found between the overall subscale score on the ACCESS and the biology ECOT test, meaning that if an EL scored high on the ACCESS, he or she would also score high on the biology EOCT. While all four of the language domains of listening, speaking, reading and writing had a positive correlation with the biology EOCT scores, writing had the greatest predictive value of the four. Analyses also included the three ACCESS composite subscale scores of oral, comprehension, and literacy. Literacy, which is made up equally of the reading and writing subscale scores, proved to have the greatest value of all of the ACCESS subscales for predicting an EL's success on the biology EOCT. Gender and length of time (LOT) in English to Speakers of Other Languages (ESOL) program were factored into a step-wise regression analysis and found to play useful roles in predicting EL success on the biology EOCT. In regard to gender, three out of four ELs who were successful on the biology EOCT were male. After the third year in an ESOL program, LOT had a negative effect on EL achievement. While these findings are consistent with recent research, an exact explanation for this phenomenon is not available in the literature. Additional data is necessary to determine the influence of school and program models, sociocultural factors, and individual differences (ID) on student success.

Electronic Thesis or Dissertation
Educational tests & measurements, Educational leadership