Sponsoring Section/Society: ASA-SOC
Session Slot: 8:30-10:20 Monday
Estimated Audience Size: xx-xxx
AudioVisual Request: xxx
Session Title: Public Policy Using Statistcs
Theme Session: Yes
Applied Session: Yes
Session Organizer: Morton, Sally C. RAND Corporation
Address: P. O. Box 2138, 1700 Main Street, Santa Monica, CA 90407-2138
Phone: 310/393-0411, X7360
Fax: 310/451-7025
Email: sally_morton@rand.org
Session Organizer: Rolph, John E. University of Southern California
Address: Information and Operations Management Dept., Marshall School of Business, Los Angeles, CA 90089-1421
Phone: 213/740-4829
Fax: 213/740-7313
Email: jrolph@rcf.usc.edu
Session Timing: 110 minutes total (Sorry about format):
Opening Remarks by Chair - 5 minutes First Speaker - 25 minutes Second Speaker - 25 minutes Third Speaker - 25 minutes Discussant - 20 minutes (or none) Floor Discusion - 20 minutes
Session Chair: Morton, Sally C. RAND
Address: P. O. Box 2138, 1700 Main Street, Santa Monica, CA 90407-2138, X7360
Phone: 310/393-0411, X7360
Fax: 310/451-7025
Email: sally_morton@rand.org
1. Is there Periodicity in the Global Mean Temperature Series?
Address: P. O. Box 2138, 1700 Main Street, Santa Monica, CA 90407-2138
Phone: 310/393-0411, X7917
Fax: 310/451-7025
Email: john_adams@rand.org
Hammitt, James K., Harvard
Hodges, James S., University of Minnesota
Abstract: Periodic fluctuations in global temeratures have drawn interest because of their role in detection and attribution (to natural or anthropogenic sources) of climatic trends. Oscilations with various periods have been reported in the literature. The 2-7 year periodicities have been associated with the EL Nino/Southern Oscillation, but the sources of the "bidecadal" (16-22 year) and longer periodicities are more speculative; possible changes in solar radiance have been suggested. A weakness of earlier research has been the inability to quantify the statistical significance of putative periodicities. We use a non-parametric bootstrap to assess the significance of apparent ENSO-related and bidecadal oscillations in global and hemispheric land and sea temerature series. Our analysis supports the existence of cycles with periods consistent with ENSO but we cannot distinguish the apparent bidecadal oscillations from random variation with data series of the length available.
2. Surveying Individuals with Disabilities
Sampson, Allan R., University of Pittsburgh
Address: Dept. of Statistics, 512 Thackeray Hall Pittsburgh, PA 15260
Phone: 412/624-8372
Fax: 412/361-6042
Email: asampson@stat.pitt.edu
Abstract: National and local disability programs require accurate and relevant statistics concerning people with disabilities. These data are needed for program advocacy, and implementing and monitoring of new and existing programs. This talk begins with a brief history of disability surveys and a discussion of current frameworks for conceptualizing disability. Issues and problems concerning disability surveys are then examined including reliability, multistage sampling, rare populations, self- identification, and effects of factors that vary over time and environment.
3. Challenges in Designing and Analyzing Social Experiments
Address: P.O. Box 2138, 1700 Main Street, Santa Monica, CA 90407-2138
Phone: 310/393-0411, X6672
Fax: 310/451-7025
Email: robert_bell@rand.org
Ellickson, Phyllis L., RAND
Abstract: Many important public-policy decisions rely on evaluating the effectiveness of social interventions. For example, preventing drug use by adolescents is an important public health and criminal justice priority. One promising approach is school-based prevention programs, which provide a convenient way of reaching almost all adolescents. To choose successfully from among the multitude of available programs, schools need hard evidence about which, if any, prevention strategies work.However, such evaluations pose substantial statistical challenges not faced by clinical trials in trying to achieve external and internal validity. This paper describes the design and analysis strategies adopted in a longitudinal experiment conducted at 30 junior high schools in California and Oregon to evaluate a school-based drug prevention program called Project ALERT. The experiment assessed whether the curriculum reduced initiation and regular use of alcohol, cigarettes, and marijuana by adolescents?
The paper discusses steps that we took to address the following challenges: achieving full, but realistic implementation of the curriculum; testing the program in diverse environments; obtaining well-matched experimental cells; obtaining reliable measures of use; adjusting for baseline differences among experimental groups; and accounting for intra-school correlation of outcomes.
Discussant: Mosteller, Frederick Harvard University
Address: Harvard Univ Science Center 1 Oxford St. Rm 604 Cambridge, MA 02138-2901
Phone: 617-495-2583
Fax: 617-496-8057
Email: NA
List of speakers who are nonmembers: None