Current Research > Methods Core > Seminars

Methods Core - Seminars

Please note: All seminars take place at CAPS in the McKusick Conference room, unless otherwise noted. Directions to CAPS.

See materials from past seminars.

Quantitative Methods

  • Friday, May 16, 2008 - 9:30 -11 am

    Presenter: CAPS Methods Core / Seminar participants / TBA
    Topic: Discussion Forum: Best Practices for Data Management: Variable creation, Data Documentation and Storage

    • Most of our seminar participants are statisticians or data analysts, and deal closely with data. We typically generate output and need to communicate the results of the runs to the PI, or explain the data to other analysts. Some of us at CAPS have discussed ideas for best practices in the creation of variable names, labels, and formats/value labels. We have also discussed possible best practices for documenting data sets and storing data and analysis outputs for easy retrieval and we have also discussed how to best communicate results to investigators and colleagues. We would now like to expand the discussion and possibly come up with some guidelines / recommendations for data management topics that come up frequently. We hope for a lively discussion. Clearly, these recommendations are non-binding, as in many situations some of the practices are a matter of personal taste and choice.

    This topic will require some "homework" on the part of participants in that we would like you to think about the practices you and your colleagues use in your work and come prepared to share what you think are effective practices in your own data analysis work. If you have a particular subject dear to your hart that you'd like to take the lead on, please let Estie Hudes know in advance. You could also prepare a small number of slides to demonstrate your point(s).

  • Friday, June 20, 2008 - 9:30 -11 am

    Presenter: Dr. William Shadish, Professor, Founding Faculty, and Chair, Psychological Sciences Section, University of California, Merced
    Topic: Can Nonrandomized Experiments Yield Accurate Answers? A Randomized Experiment Comparing Random to Nonrandom Assignment?

    • This talk presents final analyses from a study with M.H. Clark (Southern Illinois University, Carbondale) and Peter M. Steiner (Institute for Advanced Studies, Vienna, Austria) in which participants were randomly assigned to a randomized or a nonrandomized experiment. In the randomized experiment, participants were randomly assigned to mathematics or vocabulary training; in the nonrandomized experiment, they chose their training. The study held all other features of the experiment constant; it carefully measured pretest variables that might predict the condition that participants chose; and all participants were measured on vocabulary and mathematics outcome. The analyses used covariates to create propensity scores, used Rubin’s (2001) diagnostics for balance over covariates after propensity score adjustment, used covariate-adjusted randomized results as the benchmark, and compared propensity score adjustments to ordinary linear regression adjustments. Ordinary linear regression reduced bias in the nonrandomized experiment by 84 – 94% using covariate-adjusted randomized results as the benchmark. Propensity score stratification, weighting and covariance adjustment reduced bias by about 58 – 96%, depending on the outcome measure and adjustment method. Propensity score adjustment performed poorly when the scores were constructed from predictors of convenience (sex, age, marital status and ethnicity) rather than from a broader set of predictors that might include these. We present some results clarifying the circumstances under which propensity scores might work better or worse, and conclude with implications for practice.

Qualitative Methods

  • Tuesday, May 6 - 10-11:30am

    Presenter: Gretchen Purser, PhD candidate in the Department of Sociology at University of California, Berkeley
    Topic: Diving into the Labor Pool: Methodological Reflections of an “Undercover” Day Laborer
    Gretchen's research focuses on work, urban poverty and the labor market experiences of the formerly-incarcerated. Based upon comparative ethnographic fieldwork in Oakland and Baltimore, her dissertation examines the growth, structure and workings of the formal day labor (or “labor pool”) industry. With particular attention to race and gender, her talk will focus upon the methodological challenges of, as well as the analytic insights that can be gleaned from, deep immersion in this flexible “reserve army of labor.”

Past seminars