Waylon Howard will present a seminar titled "Using Principal Components as Auxiliary Variables in Missing Data Estimation." This talk addresses the handling of missing data for the applied researcher. Howard will highlight the importance of auxiliary variables when using modern missing data techniques to recover power and accuracy from data with missingness, and then focus on an important estimation limitation when using a large set of auxiliary variables. Researchers will learn about an effective and practical approach to reap the benefits of many auxiliary variables while avoiding estimation failure.
Waylon Howard is Director of Research and Evaluation at the Kennedy Krieger Institute at the Maryland Center for Developmental Disabilities, leading and managing research and evaluation efforts to assess and report on the effectiveness of various programs and grant-making strategies.
Lisa Frehill will present a seminar titled "Transdisciplinarity and the Social Science Methodological Toolkit." Transdisciplinarity has become a desirable approach to many of the world's most pressing problems. The social sciences have much to offer the development of such approaches. Frehill will review how the social science research methods toolkit can facilitate transdisciplinary research. Examples of past projects with special attention to data analytics and visualization strategies will illustrate the role social scientists can play in bridge-building across disciplinary boundaries. 12:30, W37E Machmer Hall (ISSR Lab).
Lisa Frehill is a Senior Analyst at the Energetics Technology Center in Waldorf, Marlyand. She provides support and expertise on STEM human capital (workforce and education) policy, analytics, and program evaluation.
Kimberley Geissler will present a seminar titled "Effects of Insurance Type in Physician Information Networks." Geissler uses a dataset of health insurance claims records representing 94 percent of the Massachusetts adult population to analyze the paths by which information is shared between physicians. Using propensity score methods and regression analysis, she examines relationships among network centrality measures, insurance status, and patient outcomes.
Kimberley Geissler is a Postdoctoral Fellow in Health Economics and Computer Science at the Boston University School of Management. In her research she uses econometric techniques to explore factors affecting access to effective, high-quality health care, relying on interdisciplinary methods, including advanced empirical techniques, economic modeling, and graph and network algorithms from computer science, to answer a number of policy-relevant questions.
Boston Public Library, Rabb Lecture Hall, 700 Boylston Street, Boston, MA
Join the Massachusetts Public Health Association (MPHA) in meeting the candidates for Governor of Massachusetts and find out where they stand on transportation and smart growth issues across the Commonwealth.
A healthy transportation system is critical to advancing public health and health equity. It can support public health by:
- Promoting physical activity through walking and biking - Reducing emissions and the impact of asthma and other respiratory conditions - Reducing injuries and deaths from traffic accidents - Connecting residents to economic opportunities and resources such as healthy food stores and health care facilities.
The election of a new governor is a huge opportunity to improve the health of cities and communities across the state!
This course will introduce you to a network analysis in R, provide an as-necessary introduction to R programming and cover the basics of network analysis, including terminology; data collection/storage; and basic description. We will consider advanced topics in description and exploration, such as graphical representation and community detection. Read more here.
Participants will examine some challenges of data collection in meeting-intensive settings and issues that arise when analyzing meeting-based fieldnotes, transcripts and artifacts. Participants are encouraged to bring their own meeting ethnography challenges to the session, as there will be time to discuss research design, data collection and more. Read more here.
This workshop introduces you to MTurk and its utility in facilitating social science research. You will learn how to screen data collected from MTurk samples and learn the basics of how to create online surveys and experiments in Qualtrics, an online survey tool that can be used in conjunction with MTurk. Read more here.
This workshop will provide effective strategies to prepare a competitive application for the Graduate Research Fellowship Program (GRFP). Sponsored by the National Science Foundation, these awards provide a $32,000 annual stipend for three years to early-career graduate students. This event, intended for students who plan to submit a GRFP application this fall, will include an overview of the GRFP application process, advice from our panel of past GRFP fellowship recipients, and, by request, peer and faculty review of your GRFP draft. Note: You do not need to have a draft of your application developed in order to participate in this workshop.
Attendance is by application only. The application is available here and on the GSGS blog. The deadline to submit an application is Thursday, May 15; limited to 15 students.
This event is co-sponsored by the Graduate School's Office of Professional Development/GrantSearch for Grad Students and by ISSR. Questions? Email GrantSearch for Grad Students at firstname.lastname@example.org.
Social science researchers are often interested in retrieving data from the internet. Often, this can be accomplished using so-called "web-scraping" techniques. This course introduces the statistical programming language R, then demonstrates techniques for web scraping using R. Finally, tools for analyzing text data in R will be presented. More information.
R is a free, open source programming language that gives empirical researchers a powerful set of tools for regression analysis. Workshop participants will learn how to import and export data, perform exploratory data analysis, run multiple regressions, conduct hypothesis testing, and estimate fixed and random effects models in the R environment. More information.