The Effect of a Middle School STEM Initiative on Cognitive and Non-Cognitive Outcomes: A Quasi-Experimental Study using Instrumental Variables and Inverse Propensity Weights.
Year: 2017
Type: article
Study design: Observational Study
Abstract: A Quasi-Experimental Study using Instrumental Variables and Inverse Propensity Weights. (Under the direction of Stephen R. Porter).Science, technology, engineering, and mathematics (STEM) education is an instructional framework designed to increase students' engagement and achievement in the STEM disciplines (Anderson, 2010;Bybee, 2013).This study is an evaluation of a STEM initiative in Wake County Public School System in North Carolina.The purpose of this study is to estimate the causal effect of the STEM initiative on middle school students' cognitive and non-cognitive outcomes using instrumental variable analysis and inverse propensity weights.It was hypothesized that the STEM initiative would increase students' cognitive outcomes (i.e., state standardized test scores in science, mathematics, and language arts) and non-cognitive outcomes (i.e., grit, civic engagement, future goals and aspirations, control and relevancy of coursework, peer support, and attendance at school).Grade 8 outcomes were examined near the end of the 2014-15 and 2015-16 school years for 15,840 students who either attended middle schools implementing the STEM initiative or business as usual during Grades 6-8.Results indicate that students' science achievement was lower at STEM initiative schools (d = -.19) with statistically significant effects for students who were female (d = -.18),male (d = -.20),Hispanic (d = -.18), and AIG (d = -.26), which was not a desired consequence of the initiative.The STEM initiative contributed to a decrease in unexcused absences (d = -.03) with a larger effect size for students who were limited English proficient (LEP) (d = -.23), which suggests that students may have been more engaged with school at treated schools (Holt & Gershenson, 2015).Otherwise, moderate to large effect sizes were not observed for students who belonged to subgroups at STEM initiative schools with four exceptions.At STEM initiative schools: (a) LEP students' reading achievement was higher (d = .15);(b) Black students indicated more grit (d = .20);(c) Hispanic students indicated more control and relevance of coursework (d = .21);and (d) AIG and White students indicated decreased civic engagement (d = -.23 and d = -.12 respectively).Three schools were identified as medium or high implementing schools based on interviews with the school-based STEM coordinator and school-related documents.Students at these schools demonstrated higher reading achievement and increased attendance as measured by unexcused absences, but the effect sizes were small (d = .00and d = -.02respectively).At medium or high implementing schools, four outcomes exhibited moderate effect sizes, showing that STEM initiative students demonstrated: (a) lower mathematics achievement (d = -.28);(b) lower science achievement (d = -.16);(c) lower future goals and aspirations (d = -.20); and (d) lower control and relevance of coursework (d = -.25).Collectively, these results suggest that the STEM initiative in middle schools contributes to higher student engagement as measured by attendance and reading achievement.Students at medium or high implementing treated schools demonstrated lower mathematics and science achievement, which suggests that teachers may use the instructional components of the STEM initiative (e.g., project-based learning, "the 4Cs", community and industry connections) without careful attention to accurate content.The initiative was highly contextualized at each school, making it difficult to study the impact of the CHAPTER 1: INTRODUCTION Science, Technology, Engineering, and Mathematics (STEM) education is a major focus of recent education reform efforts in the U.S. (U.S. Congress Joint Economic Committee, 2012).This emphasis is the result of an increased demand for individuals educated in STEM fields (Bybee, 2013).Between 2004 and 2008, the employment rate rose annually by an average of 3.3% in science and engineering fields compared to 1.3% in all other occupations (U.S.Department of Labor, 2007).The supply side of the pipeline reports a serious shortage of U.S. students pursuing STEM degrees (Fox & Hackerman, 2003;Schultz et al., 2011).The U.S. ranks 29 th of 109 countries in the percentage of 24 year-olds with a mathematics or science degree (Anderson, 2010).Students in the U.S. demonstrate lagging performance in STEM areas compared to students in other countries.In 2011, as measured by the National Assessment for Educational Progress (NAEP), only one-third of Grade 8 students in the United States were proficient in science (Loney, 2014).Women, minorities, and students from low socio-economic status are underrepresented in STEM fields (Anderson, 2010;Anderson & Kim, 2006; National Academies, 2005; National Science Foundation, 2010).In 2014, women held 50% of jobs in the U.S. economy, yet only 24% of U.S. women were employed in a STEM-related field (Loney, 2014).Minority students and students from low socio-economic backgrounds lag behind their White, more affluent peers in their interest in mathematics and science.Only 9% of the nation's scientific and engineering workforce included individuals who identified as Black, Hispanic/Latino, or Native American, although these groups made up 30% of the U.S. population (Anderson, 2010).Compared to
Author: Elizabeth L. Adams
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3197807818