Support

Support Options

  • Knowledge Base

    Find information on common questions and issues.

  • Support Messages

    Check on the status of your correspondences with members of the QUBES team.

Contact Us

About you
About the problem

CATALST - Change Agents for Teaching and Learning Statistics

By Joan Garfield1, Robert delMas1, Andrew Zieffler1, Allan Rossman2, Beth Chance2, John Holcomb3, George Cobb4

1. University of Minnesota 2. California Polytechnic State University, San Luis Obispo 3. Cleveland State University 4. Mt. Holyoke

View Link (HTM)

Licensed according to this deed.

Published on

Abstract

From the project website: "This project uses the acronym CATALST (Change Agents for Teaching and Learning Statistics) to represent the goal of accelerating change in the teaching and learning of statistics. The changes we are working towards are in both content and pedagogy. Our focus is the first, introductory, non-calculus based statistics course. This project was funded from 2008–2012 to develop curriculum materials, lesson plans, and corresponding student assessments."

"The CATALST curriculum consists of three units: (1) Chance Models and Simulation, (2) Models for Comparing Groups, and (3) Estimating Models Using Data. Activities are built on ideas of modeling and simulation, with “the core logic of inference” as the foundation (Cobb, 2007, p. 13). When applied to randomized experiments and random samples, Cobb refers to this logic as the “three Rs”: randomize, repeat, and reject. The CATALST project generalized this logic for a broader simulation-based approach to inference as follows:

  • Model: Specify a model that will generate data to reasonably approximate the variation in outcomes attributable to the random process—be it in sampling or assignment. The model is often created as a null model that may be rejected in order to demonstrate an effect.
  • Randomize & Repeat: Use the model to generate simulated data for a single trial, in order to assess whether the outcomes are reasonable. Specify the summary measure to be collected from each trial. Then, use the model to generate simulated data for many trials, each time collecting the summary measure.
  • Evaluate: Examine the distribution of the resulting summary measures. Use this distribution to assess particular outcomes, evaluate the model used to generate the data, compare the behavior of the model to observed data, make predictions, etc."

Cite this work

Researchers should cite this work as follows:

Tags