Volume 2 - Issue 1 - DBU Journal of K-12 Educational Research - Page 54

52 would be found by increasing the sample size of the STEM group. To accomplish this the duplication method was used where the data points are duplicated in order to increase the sample size and thereby test the mathematical model of effect size and sample size. A trend in the increase of effect size and statistical signifi - cance was found in all cases. As a result of the aforementioned analysis, the researcher noted that a significant difference in mean SAT scores would have been found between all three groups in 2015 and 2016 includ - ing Hispanic students in traditional programs, STEM programs, and choice programs if the data sets were larger. Additionally, a significant difference in mean HSGPA in 2014, 2015, and 2016 would have been found between Hispanic students in traditional programs and Hispanic students in STEM programs with larger samples. ACT data was not analyzed due to the small sample size. Additionally, the researcher conducted statistical analysis to determine if a difference existed between the mean SAT and ACT scores of Hispanic students seeking STEM endorsement at the campus in the current study and Hispanic students in Texas and the nation. The study used ex post facto data from 2014, 2015, and 2016. A significant difference was found in mean SAT scores across all three years in the study. ACT data was not analyzed due to the extremely small sample size. Table 1 displays mean SAT scores for the campus being studied, the state, and nation. Additionally, statistical significance is indicated and the mean difference in scores between the campus and the state and nation. The researcher conducted additional analysis of scores ranging over a period of three years to determine if a difference existed in SAT and ACT scores and HSGPA between Hispanic students in choice programs, including STEM, and Hispanic students in traditional programs in the district being studied. Results were mixed over 2014, 2015, and 2016. Implications An achievement gap exists between Hispanic students and White students and research has shown Hispanic students score lower on standardized tests and college entrance exams (Karantinos, 2009, Rojas-Lebouf, 2010). The achievement gap has been highly researched and noted over the past decade and does not show significant signs of improvement (Harvey, 2013). The impact of the gap has been seen in the performance of Hispanic students on the SAT, ACT, and in cumulative HSGPA (Harvey, 2013). As a result, Hispanic students are less likely to be admitted into college (Walpole, 2005). The current study indicates Hispanic students benefitted from enrollment in STEM programs. The findings from Research Question 2 indicate SAT scores for Hispanic students in the study were well above the state and national average. While further research would need to be conducted, it may be implied that the students in the current study may show sufficient growth to support some closure in the achievement gap. Additionally, His - panic students in the STEM program were found to have mean SAT scores well above college admission requirements for most 4-year universities in the state of Texas. Likewise, enrollment in choice programs and success in aca - demics indicates some Hispanic students scored higher on college entrance exams than their Hispanic counterparts in traditional Table 1. Summary of SAT Score Findings for Hispanic Students in the District and all Hispanic students in the state and Nation for 2014, 2015, and 2016. Ladye Welpman, Ed.D.