View Site Map

Carnegie Mellon University

Course Title: Introduction to Statistical Reasoning
Contact: Joel M. Smith

Status: This project originated as part of NCAT’s Pew-funded Program in Course Redesign (PCR) program, 1999 – 2003, and was successfully completed.

Descriptive Materials: In addition to the project description below, links for this project include a full academic plan, a full cost savings plan, a completed Course Planning Tool (CPT) an interim progress report, and a final project report. The final project report describes the impact of the redesign on student learning and student retention; final cost savings achieved; techniques that most contributed to improved learning and reduced costs; and, as assessment of future sustainability.

Project Plan:
Introduction to Statistical Reasoning, taught every semester, is a required course for students in Carnegie Mellon’s College of Humanities and Social Sciences and in several other majors. It enrolls between 400 and 500 students per year, over a third of the freshman class. Based on a 1991 redesign, the class now consists of two lecture meetings and one computer lab per week. The labs give students experience with designing and implementing analyses of statistical data using a statistical software package.

Employing information technology in the form of computer labs, the 1991 redesign succeeded in creating a course where students focus on the basic processes of data analysis and application of the results. The current labs and homework exercises, however, remain too prescriptive because there is insufficient time during labs and insufficient access to tutoring during homework hours to help students stay on track. CMU wants the labs and homework to be more open-ended, exploratory and active while providing more timely tutorial assistance. They also intend to change the teaching assistant staffing model to make the course less labor intensive.

The proposed redesign builds on the 1991 redesign that added computer labs. Although current lab exercises use statistical software, these are presented as written instructions for students to follow step-by-step. In the redesign, the exercises will be entirely computer-based. Computerizing the exercises will allow the introduction of the key element in the CMU strategy: an automated, intelligent tutoring system called StatTutor that will monitor students’ work as they go through the exercises, provide them feedback when they pursue an unproductive path that will lead to a mistaken or irrelevant analysis, answer questions they put to it, and closely track and assess individual student’s acquisition of skills in statistical inference.

The redesign will make labs and homework active learning exercises. Delivering more open-ended labs will be possible because StatTutor software will monitor students’ work, advise them about unproductive choices, and respond to questions that arise as they explore possible solutions to the goals they have been given. StatTutor is, essentially, an individual tutor for each student working at his or her computer. Adding the tutoring software will also reduce the number of teaching assistants required in labs.

CMU has data comparing skills of students who have taken the current course with those who have not. These data form a baseline for comparison with the results of the redesign. Formative assessment will be done in summer and fall 2000, first using a pilot class and then the first full deployment. Computerization of the labs with StatTutor will allow the team to track and analyze the paths each student takes through exercises in great detail. These data will be analyzed to see if students’ abilities to do data analysis improve as they go through the course. Special exercises will measure topic transfer skills. Finally, longitudinal studies of subsequent student course choices will be done.

This is a "second generation" redesign of a traditional lecture-plus-recitation statistics course to a lecture-plus-computer lab design. CMU’s 1991 redesign reduced costs from $227 to $195 per student. Adding StatTutor will cut the teaching assistant work force in half, reducing costs from $195 to $138 per student or an additional 29 percent savings. The total savings across the two redesigns is projected to be 39 percent.

 

 

Program in Course Redesign Quick Links:

Program In Course Redesign Main Page...

Lessons Learned:
Round 1...
Round II...
Round III...

Savings:
Round I...
Round II...
Round III...