Introduction

Mathtutor is a freely accessible web site for middle-school mathematics, where students work with artificially-intelligent tutoring software based on the successful Cognitive Tutor technology, developed at Carnegie Mellon. The Mathtutor website makes this type of computer tutoring freely available for 5 content strands for mathematics for grades 6 through 8. Mathtutor is a result of a long line of research at Carnegie Mellon University. Our goal is to share this resource with the world.

Data collection with Mathtutor has been approved by Carnegie Mellon University’s Institutional Review Board (IRB), which oversees research with human participants to ensure that it is carried out in an ethical manner.

Information We Collect and How We Use It

We use Mathtutor to collect data for three purposes:

1. to support the regular operation of the site,
2. to improve the site, and
3. to scientifically study student learning with intelligent tutoring software.

Information needed for the site’s regular operation (but not used for research) is stored in the site’s main database. Information used for research is collected separately, in de-identified form, in a central data repository called Data Shop. For some tutor problem sets, the data is also stored temporarily on one of the Mathtutor servers, from where it may be forwarded to DataShop in rare cases where logging to DataShop was unsuccessful. The temporary data store is not used for any other purposes. Only project team personnel can access the temporary data store.

How We Protect the Privacy of Users

We protect the privacy of the users of the Mathtutor site in the following ways.

1. The system’s main database is maintained on computers that only Mathtutor project personnel can access. All identifying information is encrypted. It is decrypted only for the normal operation of the site, e.g., to generate reports requested by teachers about their students or for troubleshooting technical issues.
2. Teachers will only have access to data about students in their own classes. They will not be able to access data from other users.
3. Students will only be able to view information pertaining to their own work, assignments, and performance. They will not be able to access data from other users.
4. The information in the research database (DataShop) will have no identifying information for students, teachers, and schools. It has only anonymous identifiers, for schools, classes, teachers/tutors, and students.
5. Data sets in Data Shop will be password protected, and access will be strictly controlled. The data will be made available only to educational researchers who request access. These researchers will request their own IRB approval.