|Title:||Drawing Connections to Close Achievement Gaps in Mathematics|
|Principal Investigator:||Hong, Guanglei||Awardee:||University of Chicago|
|Program:||Cognition and Student Learning [Program Details]|
|Award Period:||4 years (08/01/2017 – 07/31/2021)||Award Amount:||$1,388,030|
Co-Principal Investigator: Lindsey Richland (UC Irvine)
Purpose: The aims of this project are to examine the role of pressure as a factor underlying persistent achievement gaps in mathematical reasoning skills and to test a malleable factor for reducing those gaps: making worked examples visible while engaging students in mathematical reasoning. Most of the research exploring the effects of anxiety and pressure on academic performance focuses on testing situations. Through this project, the research team will extend this work by looking at whether anxiety and pressure impair the depth and quality of higher-order thinking (HOT) and learning as well as exploring the potential of using visible worked examples to counteract the effects of anxiety and pressure.
Project Activities: The research team will conduct one study during Years 1 and 2, one study during Years 3 and 4, and one study during the 2020-2021 school year that focuses on the COVID-19 pandemic*. In Study 1, the research team will manipulate performance pressure (i.e., pressure to perform at a high level) to look at the effects of anxiety and pressure on learning. In Study 2, the research team will manipulate both performance pressure and whether or not students will be able to use visible worked examples during learning to explore the potential of using visible worked examples to counteract the effects of anxiety and pressure. Study 3 will be conducted during the 2020-2021 school year to understand how COVID-19 related stress and anxiety affects students' learning experiences and to test whether visible worked examples counteract those effects.
Products: Researchers will produce preliminary evidence of the effects of pressure and anxiety on learning as well as potentially promising practices for counteracting those effects. They will also produce peer-reviewed publications.
Setting: Participating schools are located in diverse urban or suburban districts in Illinois and California.
Sample: Study 1 will include 14 fifth and sixth grade classrooms (approximately 330 students). Study 2 will include 18 fifth and sixth grade classrooms (approximately 430 students). The student body from which this sample is drawn has high levels of free and reduced lunch eligibility. Participating schools have high proportions of minority students, ranging from 69-85% low-income students, .5%-97% African American, .5-90% Hispanic/Latino, and 1-23% White. For the study focused on the COVID-19 pandemic during the 2020-2021 school year, participants will be 360 fifth and sixth grade students from twelve classrooms. The research team will draw participants from the student body, including 50-90% of students who receive free and reduced lunch and who are majority Hispanic/Latino.
Intervention: Due to the exploratory nature of this research, there is no intervention. Instead, this project will examine the role of pressure as a factor underlying persistent achievement gaps in mathematical reasoning skills and will test a malleable factor for reducing those gaps: making worked examples visible while engaging students in mathematical reasoning.
Research Design and Methods: The research team will conduct three studies, which are designed as randomized controlled trials in which students within each participating classroom will be randomly assigned to conditions. Participating students in each study will individually interact with a computer program where they interact with videotaped clips of real classroom lessons. These lessons offer rich opportunities for drawing connections and mathematical reasoning. Prior to instruction, the research team will administer pre-tests of students' conceptual and procedural mathematics knowledge as well as individual difference measures. Researchers will measure state anxiety, HOT, and situational interest during the lesson. The research team will administer conceptual and procedural knowledge measures again immediately after the lesson as well as after a one-week delay. In Study 1, researchers will randomly assign students to one of three conditions: performance pressure prompt before the lesson, performance pressure prompt before post-testing, or no performance pressure prompt. In Study 2, the team will randomly assign students to one of four conditions: pressure before the lesson/visible worked examples, pressure before post-testing/visible worked examples, pressure before the lesson/verbal worked examples, pressure before post-testing/verbal worked examples. In Study 3, the research team will randomly assign students to one of two conditions: math lessons that include visible worked examples and math lessons that do not. Researchers will also assess individual differences in prior math knowledge, executive function skills, and worries related to COVID-19, as well as students’ levels of anxiety and enjoyment during the lesson.
Control Condition: In Study 1, the control condition does not receive the performance pressure prompt. Study 2 is a 2 (pressure before instruction, pressure before post-test) x 2 (visible worked examples, verbal worked examples) design, so there are comparison conditions, but there is no designated control condition. In Study 3, the control condition does not receive visible worked examples.
Key Measures: Key measures include researcher developed measures of procedural and conceptual mathematics knowledge, the Flanker test of attention and inhibitory control, the d2 Test of Attention, the Symmetry span measure, the Suinn Mathematics Anxiety Rating Scale, Elementary Form (MARS-E), the Ravens Matrices task, the Patterns of Adaptive Learning Scales (PALS) - scales of Personal Achievement Goal Orientations, state mathematics anxiety ratings, a researcher developed measure of higher-order thinking, and the Situational Interest Scale. Study 3 will also include surveys to measure COVID-19 related levels of anxiety and worry.
Data Analytic Strategy: The research team will use correlation, regression, causal mediation, and moderation techniques. In addition, researchers define individual differences measures as pre-treatment predictors.
Begolli, K.N., Richland, L.E., Jaeggi, S. M., Lyons, E., Klostermann, E., Matlen, B. (2018). Executive Functions in Learning Mathematics by Comparing Representations: Incorporating Everyday Classrooms into the Science of Learning. Thinking and Reasoning, 24, 280-313.
Doumas, L.A.A., Morrison, R.G., Richland, L.E. (2018a). Individual Differences in Relational Learning and Analogical Reasoning: A Computational Model of Longitudinal Change. Frontiers in Psychology: Cognitive Science, 9:1235. doi: 10.3389/fpsyg.2018.01235
Doumas, L.A.A., Morrison, R.G., Richland, L.E. (2018b) Cognitive Control as an Underpinning of Relational Reasoning from Diagrams. Lecture Notes in Computer Science.
Frausel, R., Simms, N., Richland, L.E. (2018), Working Memory Predicts Children's Analogical Reasoning. Journal of Experimental Child Psychology, 166, 160–177.
Lyons, E., Simms, N., Begolli, K.N., Richland, L.E. (2018), Stereotype Threat Effects on Learning from a Cognitively Demanding Mathematics Lesson. Cognitive Science. 1-13,10.1111/cogs.12558.
Matlen, B. Richland E. Richland, Klostermann, E. & Lyons, E. (2018), Impact and Prevalence of Diagrammatic Supports in Mathematics Classrooms. Lecture Notes in Computer Science.
Morsanyi, K., Prado, J., Richland, L.E. (2018), Editorial: The Role of Reasoning in Mathematical Thinking. Thinking and Reasoning, 24, 129-137.
Simms, N., Richland, L.E. (2019). Generating Relations Elicits a Relational Mindset in Children. Cognitive Science, 43, e12795.
* In September 2020, $300,492 in supplemental funding was provided to support Study 3.