Boosting Physics Success with Math Support

Physics can be tricky, and it gets even trickier when you throw in the mathematical skills needed to understand it. Many students taking introductory physics courses arrive with different levels of math knowledge. Unfortunately, that often means some students get left behind, especially those from underrepresented backgrounds.

To tackle this problem and give everyone a fair shot, two clever ideas were put to the test: offering optional math assignments with extra credit and supplying AI-generated hints during homework. The goal was to see if these would help students perform better on their exams, particularly those who typically struggle.

#The Problem

Many students come to college without having taken advanced math classes like trigonometry or calculus. This situation is linked to issues like race and economic status, meaning some groups are at a disadvantage right from the start. The pandemic only worsened this situation, as remote learning made it even harder for students to keep up with math skills vital for success in physics.

This study focused on helping students who might not have had all the prep they needed by offering them more math practice and support through AI. The idea was to see if giving students these resources would make a difference in their exam scores.

#Understanding the Theory

At the heart of this study is a concept called Expectancy-Value Theory. This theory suggests that students are more likely to engage with an activity if they believe they can succeed and if they think it will benefit them in the future. In simpler terms, if students feel confident and see value in what they are learning, they are more likely to stick with it.

Two key strategies were developed based on this theory:

1. Incentivized Math Assignments: Students were given extra credit for completing supplemental math assignments, especially aimed at those who needed it the most.

2. AI-Generated Hints: Instead of a teacher always being available, hints generated by artificial intelligence were embedded in the homework assignments. This meant students could get help without the fear of judgment from their peers or instructors.

#The Study Design

The study took place at a public university where two classes of introductory physics students participated. All students had access to the same materials, but one class received the AI hints and the incentivized math assignments. The researchers closely monitored how these supports affected Exam Performance.

#Optional Supports

The first support, supplemental math assignments, focused on vital math concepts related to physics. Four main topics were covered:

1. Vectors
2. Derivatives
3. Integrals
4. Multiple integrals

Each of these math topics was designed to help students grasp the math they would encounter in their physics exams.

The second support system involved AI-generated hints for physics homework. Whenever a student got stuck on a question, they could request a hint that nudged them in the right direction without giving away the whole answer. This setup aimed to build students' problem-solving skills while reducing the stress that often comes with asking for help.

#Who Participated?

A total of 382 students participated, split between two sections of the physics course. Demographic data was collected to understand how different students responded to the supports. This included looking at factors like race, gender, and prior math preparation.

#Key Findings

#Finding 1: Increased Completion Rates

The first big takeaway was that offering extra credit significantly boosted completion rates for the math assignments. Students who were incentivized completed more assignments than those who received no incentives. This was especially true among students from underrepresented groups, who often lagged in assignment completion without the extra credit.

#Finding 2: Better Exam Performance with Math Assignments

The analysis showed that students who completed the supplemental math assignments scored better on exams. When students practiced the math topics covered in the assignments, they were more likely to perform well on the exam questions that aligned with those topics. It became clear that practice made a noticeable difference.

#Finding 3: AI Hints Helped, But Only In Certain Cases

Students who used the AI-generated hints performed better on exams when the content of the exam questions matched the homework. In one exam, students who utilized these hints saw an improvement in their performance, especially those who came into the course less prepared. However, if the exam questions didn't align well with the homework, the hints didn't seem to provide that extra boost.

#Finding 4: Closing the Gap

An important part of the findings related to disparities in performance between different Demographic Groups. When students from underrepresented racial groups completed the supplemental math assignments, they showed notable improvements in exam scores. This means that the math practice particularly helped lift the performance of students who typically face more challenges in physics.

#The Importance of Equity

The study underscored the importance of making math practice accessible to all students, particularly those from historically underserved backgrounds. By offering incentivized assignments and AI support, educators can help close the performance gaps often seen in higher education. This approach can lead to more equitable outcomes for all students.

#Looking Ahead

While the study achieved some promising results, the researchers acknowledged a few limitations. The completion rates for the supplemental math assignments were still below 60% despite the incentives. To improve this, they suggested incorporating these assignments into class time or sharing study results with future students to enhance the perceived value of the tasks.

The researchers are also eager to explore students' views on why they chose (or didn't choose) to use the optional supports. Gathering qualitative data through surveys and interviews will provide deeper insights into student experiences and motivations.

#Conclusion

This study illustrates that small tweaks in how we support students can make a considerable difference in their academic success. By combining supplemental math assignments with AI-generated hints, educators can create a more level playing field for students in physics courses.

So, if a little bit of extra credit and some friendly AI nudging can help students grasp the math needed for physics, why not give it a go? After all, who wouldn't want to ace that exam and impress their friends with their newfound physics prowess?