Volume 1 - Issue 1 - DBU Journal for K-12 Educational Leadership - Page 66

64 campuses in the district who fit the following criteria for in - clusion: 1. Dual language students who were enrolled from kindergarten, first, or second grade through the fifth grade at either dual language campus. 2. Monolingual students who were enrolled at the campus during their fifth grade year and enrolled in the district from kindergarten, first, or second grade through fourth grade. 3. Students who fit the previously stated criteria in 1 or 2 and also remained enrolled and were not withdrawn for more than 4 weeks at any time. After examination of Public Education Information Management System (PEIMS) enrollment records, there were 478 students in reading and 487 students in math who met the criteria for inclusion in the study. For the purpose of the study, the data from Campus A and Campus B were combined based on the three designated groups, and sta- tistical tests were completed for each research question and hypothesis. It should be noted that any student who exited limited English proficient (LEP) previous to the fifth grade was coded as a non-ELL student. Summary of Findings and Interpretation of Results The first question asked if there was a statistical difference in STAAR reading and math scale scores between ELLs in dual language, non-ELLs in dual language, and students at the dual language campuses who receive all instruction in English. In 2012 and 2014, non-ELLs achieved a significantly higher STAAR scale score in reading, than those students in all English. Students in all English achieved a significantly higher scale score than ELLs. In 2013 non-ELLs and students in all English achieved a significantly higher scale score than ELLs. There was no significant difference between non-ELLs and those in all English. For 2012 and 2014 on the math STAAR, non-ELLs per- formed significantly higher than ELLs and those students in all English where there was no significant difference. The 2013 math STAAR data revealed that ELLs in dual language performed significantly lower than non-ELLs in dual and students in all English where there was no significant dif - ference. After examining all six ANOVA’s, non-ELLs were the most successful students on the dual language campuses over ELLs and the students who receive all their instruction in English with the highest mean scale score on the reading and math STAAR test in 2012, 2013, and 2014. In 2012 and 2014 there was a statistically significant difference between ELLs and the all English group. When comparing the three student groups, it was expected by the researcher that the non-ELL group would achieve a higher result than ELLs. The results of the students in all English were helpful in un - derstanding the impact of the dual language program at the two campuses and are reflected in Figure 1. (See Figure 1 on page 67.) The second research question sought to determine if there was a statistical difference on STAAR reading and math scale scores between ELLs in dual language in com - parison to the local district’s mean for students who are en - rolled in the district bilingual program. In 2012, there was no significant difference between ELLs and the district mean for students in the district bilingual program. However, in 2013 and 2014, there was a significant difference noted, with ELLs in dual language performing lower than the district mean. One-sample t-tests were also performed with math data from 2012 where there was a significant difference in STAAR scale scores where ELLs in dual language performed at a higher rate than the district mean. In 2013, students in dual language (ELLs) performed significantly lower than the dis - trict mean on the math STAAR and in 2014 there was no significant difference between ELLs and the district mean. The three years of data for the second research question is inconsistent with no identifiable patterns. In addition, the ELL student group from dual language was particularly small leaving questions regarding the validity of the results (Math N = 20, reading N = 24). The third and final research question sought to compare the differences between non-ELLs in dual language with the district mean of non-ELLs who receive all of their instruc - tion in English as measured by their scale score from the reading and math STAAR. In 2012 and 2014 non-ELLs in dual language achieved a significantly higher scale score in reading and math as compared to the district mean. In 2013, Shannon Cole, Ed.D.