Project Description

 

We are developing a computer-based intelligent tutoring system that will tutor middle school students in scientific process skills aligned with current NSES inquiry skills and NAEP science process skills. We will do this for Physical, Earth, and Life Sciences, aligning content goals with those of the Massachusetts Curricular Frameworks. By process skills we mean: collecting and interpreting data, prediction- and hypothesis-making, experimenting with interactive models, mathematizing with data, and defending and communicating a scientific argument. These skills, developed in rich scientific contexts, need to be developed and assessed within the contexts in which they are embedded (Mislevy et al., 2002).

Our project, builds on extensive prior and work currently underway by the project investigators. Specifically, we will leverage:

  1. the existing authoring functionality in the ASSISTments project for Math (developed by Co-Investigators, Neil Heffernan & Ken Koedinger),
  2. the logging functionality of the ASSISTments system in order to capture students’ fine grained actions with models (Gobert et al, 2006; Horwitz et al, 2007),
  3. the existing reporting infrastructure to report students’ process skills to teachers for formative assessment so s/he can determine which skills his/her students are performing poorly on, and
  4. analyses methods for log data developed by Co-PI Ken Koedinger and colleagues at the Pittsburgh Science of Learning Center.

Science process skills can be measured using fine-grained logging technologies such as the one used by ASSISTments. The presupposition is that by engaging students in deep scientific experimentation with microworlds, and by reacting to their learning strategies in real time, we can tutor them on these strategies, and positively affect both their science process skills (as evidenced by more systematic behavior in their log files) and their content knowledge (as measured by content gains). We also expect this will significantly impact their science MCAS scores because we will align our activities and assessments to focus on the state frameworks for middle school science in Earth Science and Life Science. We will also test for transfer to other MCAS items, which utilize science process skills as well. We will use interviews, think aloud protocols, and log files in the development phases of the project. We will test our intervention in a series of randomized controlled studies (for each content area addressed) in which the control groups will receive the use of ASSISTments (i.e., the same microworlds without any inquiry tutoring) and the intervention groups will receive ASSISTments with out inquiry tutoring for each of the inquiry skills addressed. Key outcomes include the ASSISTments system for each domain, as well as empirical data regarding the efficacy of our system at improving students’ science learning across several dependent measures in each content domain.
We will be working with middle school students and teachers in the Worcester, MA public school system as well as schools in the surrounding area. Worcester is one of the largest cities in Massachusetts, with high levels of children on free- or assisted-lunch programs. The area also is home to children of many different ethnicities, and as such, provides data that are likely to generalize well across the United States

Comments are closed.