A Hierarchical Linear Modeling Approach to Understanding the Role of Ethnicity and Socioeconomic Status on Precollege Engineering Conceptions Research to Practice

Traditional precollege formal education includes state and federal mandated science and mathematics content. Most recently standards also include engineering content to support initiatives that. prepare more of the American population for the engineering challenges of the future. This study focused on a precollege engineering education intervention. Potential interactions between student and school level factors and students’ pre-test achievement were explored using a multilevel modeling data analysis approach (i.e. investigating students within schools). Findings suggest that the statistically significant predictors of the students’ pre-test scores are school socioeconomic status and ethnicity. Students who attended a school with a higher proportion of students on free and reduced lunch (FRL) – where FRL is used as a measure of socio-economic status – on average have lower scores than those who attend schools with a lower proportion of FRL receiving students. The second finding is that on average, African American and Hispanic students earn fewer points on the pre-test than do students belonging to other ethnic groups. The findings further suggest that family and community knowledge can influence student knowledge and test scores. As engineering education researchers and practitioners, we must apply these insights to the ways that we engage with diverse students and to the design of our curricula. Introduction Engineering education researchers, educators and school districts have increased interest in engineering curriculum specifically developed for elementary students, in part, because of the recent the Next Generation Science (NGSS) Standards. Prior to NGSS, forward thinking educators across the nation had already found ways to engage students in engineering thinking and practice in their classrooms. However, engineering knowledge and practice is not restricted to classroom experiences. Students often have knowledge and perceptions about what engineering is and what engineers do that is informed from their out-of-class experiences. In order to design effective curriculum and pedagogies, we must seek to understand how these conceptions are formed through lived experiences, how they impact student learning, and how they manifest in the lives of children from diverse backgrounds. Student and community socioeconomic status (SES) can also impact knowledge and learning. Environmental factors, such as family and neighborhood, can positively or negatively influence the academic achievement of students from lower SES communities. Students from higher SES communities are less impacted by these environmental factors. [3] Berliner argues that all too frequently, educational reform fails to consider the impact of SES on the student academic achievement, further perpetuating the layered issues that correlate low SES with lower the academic achievement. [3] SES plays a larger role than many educators and researcher are ready to admit. As engineering educators and researchers, we must not only focus our attention on improving interventions and instruments but we must also critically evaluate the ways that we are addressing issues of poverty and community structure in our curriculum design. Interventions and curricula must not be “one-size-fits all” rather our designs must take into consideration, the condition of the educational system and students that we are aiming to impact. We need to understand where learners are coming from when they enter the classroom. What are their experiences and how has that impacted their knowledge? What role does poverty place in their experiences? How does the SES of their community impact learning opportunities and experiences provided to the students? Not all students come into engineering with the same exposure. We must know what our students’ conceptions of engineering are and learn their norms and values. This understanding can help inform curricula design that will be aligned with student values and will make a lasting impact on their understanding of engineering and science. Literature Review Dyehouse, Yoon, Lucietto, Diefes-Dux [4] investigated the effects of the teacher development program partnership and the impact on student science and engineering learning. The results affirmed that well designed professional development could have an impact on student knowledge and engineering identity. In order to measure the effects, data included a treatment group and a control group, in which both groups took the pre-posttests at the beginning and end of the school year. These tests were administered by the teachers and then given to the researchers for analysis. Overall, the researchers found that students in the treatment group scored higher than that students in the control group on the knowledge tests and on the engineering career subscale; however, on the identity tests they found no significant difference. The researchers were also specifically examining the effects of group and sex on the post test scores. With respect to sex, the researchers found no significant difference between male and female students on any of the scales. Other studies have investigated the impact of well-designed professional development on engineering knowledge learning. Practitioners and researchers question if students are able to retain information learned from interventions. Tafur, Douglas, Diefes-Dux observed students for third and fourth grade students for two years and found that they were able to earn the highest test scores and demonstrate retention of the engineering knowledge learned in previous years. Furthermore, some second grade students who were exposed to engineering curriculum, matriculated to third grades classes that did not include engineering curriculum. When tested, these students demonstrated an increase in engineering knowledge. This was evidence that they retained the information taught to them in the second grade. Douglas, Wiles, Yoon & Deifes-Dux performed a case study on one school in the data set and interviewed four teachers about their professional development. Specifically, they were asked about their perceived value of teaching engineering in the elementary curriculum, about the support they receive to teach engineering and how they implemented the curriculum. The researchers also analyzed the students’ test scores for their respective teachers. Again students in the third and fourth grade had the highest gains in the engineering knowledge tests. Additionally, the teachers saw value in teaching engineering to young students, in areas such as teamwork and social development, in addition to learning engineering concepts. In a separate study using the same data set, the role of student sex was also explored in relation to pre-post test data on the engineering identity scale. Douglas, Mihalec-Adkins & Diefes-Dux [8] used MANOVA to analyze differences between boys and girls scores on the engineering identity pre-posttest, at each grade level. With respect to the pre-test data, the researchers expected to find a significant difference in engineering career identity by sex. However, the results from their study showed that not only did male and female students perform similarly on the pretest engineering careers identity scale but also both groups significantly improved their scores on the post-test. Another important factor, is the role of socioeconomic status and its impact on the test-scores in this data set. Douglas, Yoon, Tafur & Diefes-Dux [9] explored the relationship between a school’s socioeconomic status, teacher engineering experience, students gender, students race/ethnicity, and student prior exposure to engineering using path analysis. Key findings relative to this study include: ● Students attending lower socio-economic status schools did not perform as well as students who do not attend lower socioeconomic status schools on the science and engineering posttests, and ● Prior engineering exposure led to significantly higher pre-test scores but did not have a significant difference on posttest scores. Engineering knowledge in elementary school students is influenced by both internal and external factors. However, external factors have a disproportionate impact on the students from lower socioeconomic status communities. Underrepresented minorities are often overrepresented in these communities. This study addresses the intersection between the projected growth of underrepresented minorities in the United States, the disproportionate representation of minorities in lower SES communities, the role of SES on academic achievement and the need to prepare Americans for future engineering challenges. The purpose of this study is to examine the school and personal factors that are significant to elementary students’ understanding prior to classroom experiences with engineering. The following research questions guided this investigation: ● When factoring in SES, how does that change our understanding of the ways we teach engineering to elementary students? ○ What is the impact of SES on elementary students learning engineering? ○ Do school characteristics, student ethnicity, and the sex of the student help explain student scores on the 15-item Engineering Knowledge Test given to 2 4 graders? Methods Research Design Given the stated purpose of this study, the data were analyzed using multilevel modeling. In social science research, multilevel modeling is a preferred method when data is nested in different levels (i.e. children in families, teachers in schools, schools) neighborhoods. Multilevel modeling was selected for this study because other statistical analysis methods do not take into account the independence between the measured data and the levels. Hence, increasing the difficulty of measuring each levels effects on the outcome under investigation. The multilevel analysis (HLM) is briefly described in the analysis section. Setting and Sampling The data collected for this