MAPSS

EITM: http://www.poli.duke.edu/eitm/application.htm

“The scientific study of politics requires empirical evaluation of theoretical models, but theories too often proliferate without adequate testing, and empirical work too frequently applies sketchy and oversimplified theory. In EITM, researchers use recent advances in game theory and mathematical modeling to develop theoretical models of politics. These models are then subjected to rigorous tests that meet the highest standards of empirical research, including statistical analysis, experiments, and case studies. In some instances, researchers create new estimators designed to closely test the assumptions and predictions of the theoretical models. By integrating models and data, EITM is creating a new standard for theoretically grounded empirical research that yields cumulative advances to our understanding of politics.

Recognizing that gaps between theory and empirical method seriously impair scientific progress, the Political Science Program of the National Science Foundation supports annual four-week summer institutes on Empirical Implications of Theoretical Models (EITM). Previous summer institutes have taken place at Harvard (2002), Michigan (2003 and 2006), Duke (2004), UC-Berkeley (2005) and UCLA (2006).

Funding to defray participants’ costs of travel, accommodation, and subsistence is available. EITM institutes are selective, with admission based significantly on the quality and potential of research presented. Institute training includes teaching and research components, providing students a highly individualized interaction with a far wider and deeper array of mentors than is available at any individual institution. Female and minority applications are strongly encouraged.”

Andrew Vickers, a biostatistician at Memorial Sloan-Kettering Cancer Center in New York, wrote in the NY Times lamenting the reluctance to share data.

”

Dr John Kirwan, a rheumatologist from the University of Bristol in England, has studied researchers’ attitudes on sharing data from clinical trials. He found that three-quarters of researchers he surveyed, as well as a major industry group, opposed making original trial data available. It is worth restating this finding: most scientists doing research on how best to help those in pain, or at risk of death, want to keep their data a secret.

Dr. Kirwan went on to ask his subjects why. Their reasons were entirely trivial: one cited the difficult of putting together a data set (wouldn’t this have to be done anyway in order to publish a paper?); another was concerned that the data might be analyzed using invalid methods (surely a judgment for the scientific community as a whole). This is something of a clue that the real issue here has more to do with status and career than with any loftier considerations. Scientists don’t want to be scooped by their own data, or have someone else challenge their conclusions with a new analysis.”

From Torsten Neilands

Dear Colleagues,

A project on which I’m working recently received continuation funding to collect cohort data for another five year funding cycle. The project’s PI, Elise Riley, and its project director, Jennifer Cohen, are looking to hire a statistician at 50% effort. The job description appears below. Please feel free to forward this job announcement on to interested individuals or groups and please direct all inquiries about the position to Jennifer Cohen (her e-mail address appears in the job description).

With best wishes for a happy 2008,

Tor Neilands

We are seeking a statistician for 50% employment with our UCSF research study regarding the longitudinal effects of HIV, housing, drug use and social support on the victimization and risk behavior of unstably housed women.  Strong organizational skills and an ability to communicate well will be necessary in this position.  Three to five years of statistical experience are preferred and no supervisory responsibility is involved with this position.

In collaboration with the PI and the senior statistician, the incumbent will choose appropriate statistical techniques, which will guide study development as well as determine the interpretation of study results.  Prior experience with complex data sets will prepare the incumbent to develop the analyses and protocols needed in this study.  Job duties will include the following:  documenting all procedures; assessing survey question validity and reliability; performing descriptive statistics;  performing time to event and adjusted analyses such as logistic, linear and Poisson regression, Kaplan-Meier estimation, and Cox proportional hazards survival analysis; performing longitudinal analyses that adjust for correlated data such as random effects and GEE modeling techniques; conducting power analyses for all projects and grant proposals; writing the first draft of results sections for manuscripts and analytic methods sections for grant proposals.

The incumbent must be proficient with SAS or equivalent statistical software that has the capacity to analyze longitudinal data with time-varying repeated measures (e.g., R, S+ or STATA).  In addition, the incumbent must be adept with database systems such as ACCESS.  Experience with computer-based interview systems is preferred; if the incumbent has none, s/he must be willing to become acquainted with the Questionnaire Development System (QDS) package.  Weekly check-in and/or meetings with the PI, monthly all-staff meetings, and periodic briefings will provide the incumbent multiple opportunities to assess progress, prioritize work, explore new ideas and discuss problems.

Interested applicants should contact Ms. Jennifer Cohen for more information (jcohen@epi-center.ucsf.edu) and apply through the UCSF Career website: http://www.ucsfhr.ucsf.edu/careers/ Please apply for the Statistician position, Requisition number: 24552BR.