Overview

River City is a multi-user virtual environment (MUVE) consisting of a 19th century city with a river running through it, different forms of terrain that influence water runoff, and various neighborhoods, industries, and institutions such as a hospital and a university. The students themselves populate the city, along with computer-based agents, digital objects that can include audio or video clips, and their instructors' avatars. They work in teams to gather data, develop hypotheses regarding one of three strands of illness in the town (water-borne, air-borne, and insect-borne) and then to test their hypotheses.

The three disease strands are integrated with historical, social and geographical content, allowing students to have experience  with multi-causal problems embedded within a complex environment.  After testing their hypotheses, students analyze their data using graphs and tables and then write an authentic lab report on their findings in the form of a “Letter to the Mayor of River City.” Finally, at the end of the project, students compare their research with other teams of students in their class to delineate the many potential hypotheses and causal relationships embedded in the virtual environment.

Three River City variants were compared to a “control” condition that utilized a paper-based curriculum in which the same content and skills were taught in equivalent time to comparable students without using computers.  Variant GSC centers on a guided social constructivist (GSC) model of learning-by-doing, in which guided inquiry experiences in the MUVE alternate with in-class interpretive sessions led by the teacher.  Variant EMC shifts the learning model to a situated pedagogy with expert modeling and coaching (EMC) based on expert agents embedded in the MUVE.  Finally, variant LPP also uses a situated learning model but based on a community of practice.

Study Design

Study Type: Quantitative/Qualitative (Mixed Methods). 

Data collection included:

• pre and post intervention student and faculty surveys
• entries in a database that is part of the MUVE that can track student inquiry behaviors
• the student's performance on a content test
• reviews of their letters of the "Letter to the Mayor."
• semi-structured interviews with groups of students

The three variants (GSC, LPP, EMC) were assigned randomly to students within each classroom, while the control condition was assigned randomly to entire classes.

Participants:  approximately 2000 students in major urban areas in New England and the Midwest as well as one mid-atlantic suburban area; schools in these areas had high proportions of ESL and free-and-reduced-lunch pupils.  A total of 8 schools, 12 teachers and 61 classes are involved in this analysis. 

Research Questions:

1. Do students engage in inquiry in River City?
2. When compared to the “control” version, what types of significant gains in affect and learning

   for both content and inquiry do versions GSC, EMC and LPP produce?  

3. How do results on inquiry learning compare between standardized type testing and performance

   assessments?

4. What are MUVEs’ strengths and limits in facilitating classroom-based inquiry learning?

Key Findings

• data gathered from the program's database shows that students engaged in more data gathering behaviors and consulted more sources.  From visits 2-4 students' data gathering behaviors went up from an average of 12 to 16.  Over the same visits the number of sources consulted (there were 8 possible sources they could pick from) went up from an average of 2 to 4.
• interest in science careers was 5% for students in the River Walk curriculum than for the control group.
• student scores on a thouthfulness of inquiry measure were higher for the River Walk students than for the control group on the first implementation.  However, on the second and third implementations the scores depended on the pre-test scores.  For the students who scored low on this measure on the pre-test, the scores were slightly higher for the control group, but for students with average and above-average scores on the pre-test the scores were higher for the River Walk students in the post-test.
• Both the River Walk and the control group students improved in the content test scores, but the River Walk students had a bigger improvement (32-35% as compared to 17%). 
• when looking at the performance according to the different methods of evaluation:

         - on survey questions, the control group did slightly better than the GSC and EMC groups on the

            first implementation and there was an additional difference by gender for the second and third

            implementations, with boys doing better than girls for every variant except for LPP.  Another

            interesting finding is that the EMC group's scores did not vary according to their previous 

            science grades, while the other groups maintained a difference between students with low

            and high science grades.

          -

Strengths:

Weaknesses: