| Session Title: Statistical Methods for Interrupted Time-Series Analysis: Using the Auto-Regressive Integrated Moving Average (ARIMA) Technique in Program and Policy Evaluation |
| Demonstration Session 346 to be held in Santa Monica on Thursday, Nov 3, 11:40 AM to 12:25 PM |
| Sponsored by the Quantitative Methods: Theory and Design TIG and the Crime and Justice TIG |
| Presenter(s): |
| Derek Cohen, University of Cincinnati, cohendk@mail.uc.edu |
| Abstract: This demonstration is designed to present attendees with the "bare bones" essentials to using ARIMA, a sophisticated interrupted time-series statistical model. Attendees will first be briefly educated in the mechanics and assumptions of using of the ARIMA technique. Then, the presenter will demonstrate how to compile or acquire an appropriate dataset as well as how to prepare it for statistical analysis. Once the data is prepared, attendees will be shown the basics of model construction, as well as how to alter a model to minimize autocorrelation. The output will then be interpreted and explained to attendees so that they will be able to garner conclusions from their personal data and models. The demonstration will be conducted using the SPSS/PASW statistical software package. The data used in the demonstration is from a work-in-progress evaluation of firearm policy in the state of Ohio. |