Summer Research Methods Workshop
Welcome to the UB Summer Research Methods Workshop. The UB Research Methods Workshop brings together junior and advanced researchers in a collegial, intellectually engaging, and multidisciplinary atmosphere. We provide workshop participants with easy-to-understand, practical information about a variety of statistics and methods. Students do hands-on work to use their new knowledge and skills to confidently collect and analyze their own data.
Coming fall 2016
Qualitative Data Analysis
Professor Elizabeth A. Gage
This course will introduce qualitative analytic approaches, including grounded theory and thematic analysis, and guide students in applying them to data. We will discuss strategies for study design to allow robust data analysis, management of qualitative data, transcription and data preparation for analysis. The course will offer in-depth focus on code book development, coding data, and interpretive frameworks for data analysis. We will discuss training and managing teams for qualitative data analysis, inter-coder reliability, establishing consensus among coders, and computer software packages for qualitative data analysis. We will also cover techniques for writing and presenting qualitative data and methods.
Structural Equation Modeling
Professor Craig Colder
This course introduces participants to structural equation modeling (SEM) both with and without latent variables. It is designed to provide the necessary skills to use SEM to analyze social science data. The course will primarily focus on the application of SEM as opposed to mathematical underpinnings. We will cover introductory topics including path analysis, moderation (multiple group models) and mediation, confirmatory factor analysis, and hybrid structural equation models (models that include causal paths between latent variables). Special advanced topics will include working with non-normally distributed variables, missing data, categorical observed variables, and latent interactions. Upon finishing the course students will be able to estimate models using available software (Mplus) to conduct research and interpret results. The course also prepares students for more advanced topics, including latent growth curve analysis and latent mixture modeling.
Longitudinal Data Analysis
Professor Robert L. Wagmiller, Jr.
Longitudinal panel data offer many advantages over traditional cross-sectional data. Panel data not only enable researchers to better understand how individuals and their experiences change over time, but they also offer researchers better ways to evaluate cause-and-effect relationships. This course provides participants with an introduction to key methods for analyzing panel data. We will discuss fixed and random effects models for panel data, cross lagged panel models, latent growth models, and latent growth mixture models. The course will focus on the application of these models, as well as how to select the most appropriate model for a researcher’s research question. On each day, the morning session consists of lectures, and the afternoon session of a computer lab where everyone can practice. During the computer lab there is the possibility to discuss your own data.