The Center for Integrated Latent Variable Research (CILVR)
presents
ONLINE SHORT COURSE: ANALYSIS OF COMPLEX SURVEY DATA (FROM NCES), November 19-20, 2021 (Friday-Saturday)
taught by
Laura Stapleton, University of Maryland
Short Description
ONLINE SHORT COURSE: ANALYSIS OF COMPLEX SURVEY DATA (FROM NCES)
A wealth of publicly available national and international data exists for use by researchers in education and other social science disciplines. Of particular interest for this workshop are the data supplied by the National Center for Education Statistics (NCES) although the topics presented in this workshop generalize to other large-scale data collection. The goal of this workshop is to allow those researchers who are new to using national and international data to become more comfortable with accessing and appropriately analyzing the data. This short course is meant to introduce participants to the issues in working with national and international complex probability sample data sets, including both conceptual issues in measurement and setting up models as well as in the specialized statistical procedures required to conduct appropriate analyses. Participants will be presented with structured examples of downloading data and addressing the analytic challenges, as well as be given an opportunity to explore their own analyses with feedback from the instructor. At the end of the short course, participants should be able to:
- Learn about and download public-release data from the National Center for Education Statistics
- Acknowledge the limitations in using these data for analyses, given constraints in measurement and the observational nature of survey data
- Describe the differences between the many types of weights available on the data set (e.g., sampling weights, panel weights, replicate weights)
- Undertake basic statistical analysis (descriptive analysis, t-tests, multiple regression), obtaining appropriate unbiased estimates and standard errors by using sampling weights and specialized variance estimation techniques (supported software for basic analysis will include SPSS, SAS, R, Stata, and Mplus)
- Recognize advanced issues that may need to be addressed, such as imputation methods for missing data and domain analysis for studies of subpopulations
- Identify advantages and disadvantages in utilizing multilevel models with these types of data
Course Fees
Professional: $345
Full-time student*: $195
*Full-time students must submit student status proof at https://go.umd.edu/CILVR-STUDENT for prompt processing of the registration.
Free for registered HDQM Department faculty and degree-seeking students, although you must register through the internal link.
REFUND POLICY: Full refund if cancellation occurs at least 10 business days prior to the workshop date; 50% refund if within 10 days of the first day of the course.
More details and registration instructions available on the Complex-2021 page.
For any questions, please contact Patrick Sheehan at sda.cilvr@gmail.com