Math Teacher Training, Student Math Performance and Math-related Career Expectations

• In Event: Emerging pedagogies for powerful mathematics teaching

Proposal

Introduction / Significance of the Study
Investments in human capital, such as education and professional training, are essential for economic development (Becker, 1962, 2009). As the largest professional workforce in the U.S. (Goldhaber & Brewer, 1996), teachers play a critical role in shaping student academic performance, cognitive development, and career trajectories, all of which contribute to national economic growth and competitiveness (Liu, 2021; Hanushek & Rivkin, 2010). Teachers are one of the most valuable resources in education (Hattie, 2009).
However, some studies raise concerns about the effectiveness of teacher training. Yoon et al. (2007), after reviewing over 1300 studies, found only nine that showed a positive impact of teacher professional development on student achievement. Demonte (2013) also critiqued teacher training for being too generic, infrequent, and lacking subject specificity, which limits its effectiveness in improving student outcomes.
A persistent issue in the U.S. and other Western countries is "out-of-field teaching," where teachers are assigned to subjects outside their expertise (Ingersoll, 1998). This is especially problematic in mathematics, where unqualified teachers are linked to poorer student performance (Akiba et al., 2007). Despite efforts to address this, out-of-field teaching remains widespread (Barlow, 2002; Du Plessis, 2013). Research also suggests that variations in teacher quality contribute to declining student interest in STEM careers (Steidtmann et al., 2023).
Improving teacher quality, particularly in math education, has been a focus of research (Darling-Hammond, 2000). Studies highlight the importance of subject-specific training to enhance student outcomes and support economic competitiveness. Since teaching training forms the foundation of future teaching quality, it is important to explore how teacher qualifications, particularly in math, influence student performance. However, few studies have examined this relationship, and existing results are inconclusive (Ker, 2013).
This study uses U.S. data from the 2019 Trends in International Mathematics and Science Study (TIMSS) to explore how different types of math teacher qualifications affect student math achievement and future STEM expectations. In particular, we examine the following two questions:

Research Questions
1. Does having a teacher with a mathematics training background positively affect student math achievement?
2. Does having a teacher with a mathematics training background positively affect student STEM expectations?

Data and Variables
This study uses data from the U.S. sample of the 2019 TIMSS, the latest wave of the international study. TIMSS provides a nationally representative dataset of students in fourth and eighth grades. For this analysis, we focus on eighth graders, as math achievement and STEM expectations are typically more fully developed at this stage. TIMSS includes data from student assessments and questionnaire responses from teachers, school principals, and students, providing rich information on academic backgrounds, classroom practices, and school resources.
The dependent variables in this study are students' math achievement and their STEM expectations. TIMSS calculates five plausible values for each student's math achievement. To account for this, we use the pv package in STATA, which incorporates all five plausible values and estimates results with replicate weights. This ensures that the estimates are accurate, considering the structure of the plausible values.
To measure STEM expectations, we analyze student responses to three key questions related to their attitudes toward math and STEM careers:
1. "I would like a job that involves using mathematics."
2. "I need to do well in mathematics to get into the university of my choice."
3. "Mathematics is one of my favorite subjects."
Students responded to these questions on a four-point Likert scale, and we recoded their answers into binary variables (1 = agree a lot/a little, 0 = disagree a lot/a little) to reflect their STEM expectations.

Method of Analysis
To answer the first research question, we created a categorical variable representing different types of teacher qualifications. This variable includes four categories:
1. Teachers with a major in both Mathematics and Mathematics Education.
2. Teachers with a major in Mathematics but no major in Mathematics Education.
3. Teachers with a major in Mathematics Education but no major in Mathematics.
4. Teachers with other majors.
We used OLS regression to examine the impact of teacher qualifications on student math achievement. Control variables in the model include student gender, school-level socioeconomic status, teacher gender, teacher experience, number of books at home, parental education, and student immigration status.
For the second question, where the dependent variable (STEM expectations) is dichotomous, we applied logistic regression. This allows us to assess whether different types of teacher qualifications are associated with changes in students' attitudes and career aspirations in STEM.

Preliminary Findings
Table 1 presents descriptive statistics for the dependent and independent variables. The majority of U.S. 8th-grade math teachers hold qualifications in both mathematics and mathematics education. This group is followed by teachers with unrelated majors, those with qualifications in only mathematics education, and finally, those with only a mathematics major.
Table 2 summarizes the OLS results, showing that teachers with both mathematics and mathematics education qualifications significantly improve student math scores by six standard deviations compared to other majors. However, teachers with only a mathematics or mathematics education major show no significant difference compared to those with unrelated majors.
Table 3 presents the logit regression results for STEM expectations. No significant differences were found for any teacher qualification type in terms of students' attitudes toward mathematics or STEM career aspirations.

Discussion and Conclusion
These preliminary findings underscore the positive impact of teacher qualifications on student math performance, particularly for teachers with training in both mathematics and mathematics education. This highlights the importance of subject-specific training and comprehensive teacher preparation in improving student outcomes. While teacher qualifications did not significantly influence students' math attitudes or STEM career expectations, the notable effect on academic achievement underscores the value of investing in high-quality teacher education. Future research will explore whether different teaching strategies across qualification types can further explain these findings, offering insights into how teacher training can be tailored to better influence both academic performance and student aspirations.

Authors

• Huacong Liu, Shanghai Jiao Tong University

• Lechen Li, University of Florida

• Bingxin Cao, University of Florida

• Frank Fernandez, University of Florida

• Volha Chykina, University of Richmond