Exploring the factors associated with students’ experiences in middle school math classes

• In Event: Fostering Equitable, Culturally Responsive Instruction System-Wide: A Multidimensional Examination of Mathematics Education in Four Districts

Abstract

Purpose
Many educational interventions in mathematics aim to improve student achievement. While achievement is an important outcome, it does not capture other important factors, like students’ experiences. These additional measures are important for students’ wellbeing, and also because students’ beliefs about math are closely related to their future performance. Students who do not believe that they can perform well in math tend to have lower achievement than students who believe that they can (e.g., Chen, 2003; Cleary & Chen 2009; Mason & Scrivani 2004). This is particularly important for students who are Black, Latino, multilingual learners, or experiencing poverty, who are the most at risk from disengaging from school during the middle school years (Balfanz & Byrnes, 2006). This paper explores the relationship between aspects of student experience and the student-, teacher-, classroom-, and school-level conditions that may influence these experiences.

Theoretical Perspective
We posit that students who participate in rich, engaging learning environments can thrive. We define these student outcomes by 5 constructs that are both related to positive experiences and which can be improved: math enjoyment, achievement identity, performance, persistence, self-efficacy, and growth mindset (for example, Bandura, 1997; Boe et al., 2002; Burgoyne et al., 2018; Goetz et al., 2008; Lopez, 2017)

Data
We draw on survey data administered to over 5,000 students from Fall 2021 to Spring 2023. The survey included questions related to our 5 measures of student experiences, as well as basic demographic questions. We supplement these data with survey data from these students’ math teachers, data from classroom observations of the students’ math classes (analyzed for rigor and cultural responsiveness using the MSCAN and the CRMT classroom observation tools, described earlier), and publicly available data about the students’ schools.

Methods
We analyze our data using a regression framework. First, we fit a series of binary regression models for each student experience outcome using student-, teacher-, classroom-, and school-level characteristics as predictors. Next, we fit a multivariate regression model for each student outcome, using only the predictors that were statistically significant in the binary models. This additional step allows us to test the predictors that explain student experiences, after taking other variables into account.

Results
Preliminary results suggest that student demographics are strong predictors students’ experiences. On average, female students report lower levels of math enjoyment, self-efficacy, and achievement identity relative to male students, despite reporting higher levels of engagement. Moreover, students whose teachers had higher confidence meeting students’ needs reported higher levels of math enjoyment, achievement identity, and student engagement. More complete results will be presented, including how observed classroom practices are related to student outcomes.

Significance
This study adds to the literature by exploring the factors related to student experiences in math classrooms, an important yet often overlooked outcome. The results can shed light about the students who are most at risk of having negative experiences, and the factors that are most promising for improving students’ beliefs and engagement in math.

Authors

• Lauren Banks Amos, Mathematica

• Shira Haderlein, Mathematica Policy Research, Inc