Poster #86 - The Developmental Growth Trajectories of Mathematics Anxiety: A Latent Growth Mixture Model Analysis

• In Event: 3-003 - Poster Session 09
  In Poster Session: PS 09 Section - Education, Schooling

Integrative Statement

Mathematics anxiety (MA) is defined as a feeling of tension, worry, and fear experienced prior to or during math-related activities. Prior research showed that students with higher MA generally had lower math achievement, exhibited more negative math attitude, and were more avoidant of math (Ho et al., 2000).

Given the important role MA plays in mathematics development, it is imperative to identify students prone to the development of high MA and to explore possible predictors of such developmental patterns. Unfortunately, these developmental issues concerning MA were largely unstudied because most existing studies on MA rely on small samples and cross sectional data. The present study aimed to address these gaps by utilizing a nationally representative longitudinal sample of high school students from the Longitudinal Study of American Youth (LSAY; Miller, 2014). Specifically, this study 1) explored the developmental trajectories of MA presented among high school students in the U.S.; 2) examined personal and environmental predictors of these developmental trajectories; and 3) examined the educational consequences associated with each developmental trajectory.

Beginning in 1987, 5945 10th grade students (49% female) from 95 schools across the U.S. were recruited and assessed in 3 annual waves. Students’ mathematics achievement scores reflected their performance on basic math skills, algebra, geometry, and quantitative literacy. Students self-reported their MA in grades 10 to 12, persistence in challenges, as well as perception of their math teachers’ caring and strictness in grade 10 using questionnaires developed by the LSAY team. Students were contacted again in 2006 and reported their first college major.

Latent growth mixture model analyses suggested that there were 5 distinctive developmental trajectories of MA. Most students (class 4) showed stable modest levels of MA throughout high school years. About 6% of students (class 1) exhibited high MA at the beginning, but had dramatic decrease in MA over the course of high school. About 2% of students (class 2) exhibited drastic increase in MA in high school. About 3% of students (class 3) showed stable high levels of MA. Finally, 8% of students initiated with moderate levels of MA which increased slightly in high school.

Using multinomial logistic regression, it is found that lower initial math achievement predicted higher probability of being in classes 1 and 3 compared to the “norm group” (class 4). A more caring and strict math teacher predicted higher probability of being in class 2 and lower probability of being in class 3 compared to the norm group. Finally, a more persistent student was less likely to be in classes 3 or 5 compared to the norm group.

Class memberships predicted math outcomes. Specifically, classes 1, 3, and 5 had lower math achievement compared to the norm group at the end of high school. Class 2 were more likely to choose STEM majors in college compared to the norm group.

Results suggested that increases in MA in high school can be adaptive or maladaptive, and highlighted the importance of examining the heterogeneity of the developmental trajectories of MA in the student population.

Authors

• Zhe Wang, Texas Tech University
  Presenting Author

• Margherita Malanchini, The University of Texas at Austin
  Non-Presenting Author