Seminars | Courses
To help sociologists keep abreast of recent scholarly trends and developments, the Program Committee creates specialized seminars. Experts considered to be at the forefront of a given field are invited to conduct these sessions. Seminar topics and leaders are listed below. All sessions are run seminar-style; there will be NO hands-on computer work.
Attendance at each seminar is limited to 50 registrants. Prepaid registration is required; fees are $30.
Didactic Seminar. Designing and Implementing Large Scale, Comparative, Qualitative/Ethnographic Research
Sunday, August 13, 4:30pm - 6:10pm
(Leader) Kathryn J. Edin
Didactic Seminar. New Methods of Analyzing Social Networks
Friday, August 11, 10:30am - 12:10pm
(Leader) Joseph Galaskiewicz
The seminar will examine different data analytic techniques for analyzing social network data. The attendee should have a familiarity with how network data are collected, the organization of network data into matrix form, matrix algebra, and basic statistical methods. The seminar will utilize data collected in different urban settings (Chinese cities, Phoenix-Mesa, and Minneapolis-St. Paul) and will apply log linear models, correspondence analysis, and discrete choice models to these data.
One purpose of the session is to show how these different data analytic techniques can be used to study network data and to highlight some of the advantages and pitfalls of these methods. Another purpose is to focus on how to use these methods to test different substantive theories about patterns of stratification in urban communities.
Didactic Seminar. Theorizing: Interpretive Work in Qualitative Analysis
Sunday, August 13, 10:30am - 12:30pm
(Leader) Diane Vaughan
Everyone talks about theory, but few talk about theorizing: the process of explaining their data. In published work, the researcher's process of interpreting the data is seldom included so is largely invisible. To a great extent, the interpretive process remains an individually-developed skill, comprised of tacit knowledge that is difficult to articulate, to convey, and therefore to teach. In this seminar, we will examine sources of explanation in qualitative data analysis, with the goal of making the invisible visible.
Our primary focus will be on data gathered in field research using ethnography and/or interviews; we will also consider research that relies on secondary analysis and historical sources. We will explore two sources of theoretical insight and explanation: analogy and analogical comparison. Our conversation will be based on examples from published research and research experience that reflect a variety of qualitative methods and projects.
Didactic Seminar. Topics in Regression Modeling
Sunday, August 13, 2:30pm - 4:10pm
(Leader) Roger A. Wojtkiewicz
This methodological seminar is directed to experienced researchers who use regression analysis in their research. The prerequisite for the seminar is knowledge of statistics and regression analysis as taught in a one-year graduate sequence.
There are generally three kinds of hypotheses which can be addressed with regression analysis. The simplest is whether there is an effect for an independent variable on a dependent variable. This effect is easily estimated by bivariate or multivariate regression. A second kind of hypothesis is about how other independent variables, often called control variables, explain the effect of an independent variable on a dependent variable. This hypothesis is modeled by adding control variables in some sequence to a baseline model.
The third type of hypothesis is about how the effects of an independent variable on a dependent variable are contingent on the level of a second independent variable. Regression models with interaction variables address this kind of hypothesis. The purpose of this methodological seminar is to expose participants to underlying conceptual issues behind using regression modeling to address the second and third types of hypotheses.
The seminar has three main parts: 1) strengths and weaknesses of various approaches for considering the influence of control variables; 2) alternative approaches for modeling interactions involving categorical and interval variables; and 3) consideration of the additional issues of three-way interactions, splines, and linearity. Control modeling approaches to be discussed include adding control variables one-at-a-time, adding control variables in steps, and considering the influence of a control variables when other control variables are themselves controlled. The interaction segment of the seminar will consider the standard interaction model, the within-group effects model, and the ""separate model"" model.
Didactic Seminar. Bayesian Statistics
Monday, August 14, 8:30am - 10:10am
(Leader) Scott M. Lynch
Bayesian statistics is relatively new in sociology, but over the past 15 years, advances in methods to estimate Bayesian models have made Bayesian modeling of sociological phenomena more feasible. In addition, advances in the complexity of sociological models (e.g., the development and use of hierarchical models and methods for panel data) have made Bayesian statistics more desirable, and in some cases more appropriate, than classical methods.
This course is a highly applied course that begins by reviewing concepts from probability theory, explains the Bayesian perspective on probability modeling and statistics, and shows how models in sociology, ranging from simple to complex, can be estimated within a Bayesian framework using Markov chain Monte Carlo (MCMC) sampling methods. Although the primary focus is on the estimation of parameters that may also be estimated via classical methods, the course also discusses advantages of the Bayesian approach in terms of model evaluation, comparison of non-nested models, hypothesis testing, handling of missing data, and estimation of ancillary parameters.
The course will involve discussion of programming MCMC algorithms in R, and so some time will be devoted to demonstrating basics of R syntax. By the end of the seminar, participants should have a basic understanding of (1) the Bayesian perspective on probability and statistical modeling, (2) the use of MCMC methods to estimate model parameters and summarize knowledge of them, and (3) the implementation of MCMC methods via R. An understanding of classical statistical analysis using maximum likelihood estimation is assumed but will be briefly reviewed.
Didactic Seminar. Multilevel Models
Friday, August 11, 4:30pm - 6:10pm
(Leader) Peter V. Marsden
This seminar introduces multilevel regression models (also known as hierarchical linear models) for data bases in which observations on some units are nested within others. Examples include data on employees within organizations, students within classrooms, and repeated measures within individuals. Coverage ranges from elementary components-of-variance models to random-coefficient regression models. Seminar topics include model specification, estimation methods, criticism and model-checking, inference, and interpretation of findings. Emphasis is on models for continuous outcomes, but generalized linear mixed models for the study of binary responses will also be introduced. Some multilevel analysis software will be surveyed. Participants should have a good working knowledge of linear regression analysis and its extensions.
Didactic Seminar. Event History Methods
Saturday, August 12, 1:30pm - 4:10pm
(Co-Leader) Lawrence L. Wu, (Co-Leader) Jui-Chung Allen Li
This seminar will provide a brief introduction to the analysis of event history data, with an emphasis on on continuous-time models and estimation. Topics include data structures for event histories and time-varying covariates, right and left censoring, left truncation, exploratory methods, proportional hazard models, alternative specifications of time dependence, estimation, and hypothesis testing. Seminar assumes familiarity with multiple regression, inference, and basic statistical inference, but no prior knowledge of event history methods.
Didactic Seminar. Computer Assisted Software for Qualitative Data Analysis: How to Integrate Software into Your Analysis of Qualitative Data
Friday, August 11, 2:30pm - 5:00pm
(Leader) Sharlene J. Hesse-Biber
"This didactic seminar is for qualitative researchers who wish to use computer software to analyze textual data (e.g., case records, newspaper articles,fieldnotes, transcripts of interviews or focus groups discussions), pictures, graphics or audio and video tapes/discs. We will discuss the factors you should consider in selecting a software program. We provide specific examples on how to integrate software into your analysis of qualitative data using the program HyperRESEARCH. We describe how to code, memo, and retrieve your qualitative data. We will demonstrate how to use software to perform analytical induction ( grounded theory) as well as how to use software in mixed methods studies.
This is NOT a hands-on session, but you will be provided with a range of workshop materials to utilize as you take these ideas back to your own analysis and specific research projects. The seminar will run for 2.5 hours"
Didactic Seminar. Methodologies of the History of Sociology (co-sponsored by the ASA Section on History of Sociology)
Saturday, August 12, 8:30am - 12:10pm
(Leader) Edward A. Tiryakian, (Panelist) Charles Tilly, (Panelist) Craig Calhoun, (Panelist) Jack A. Goldstone, (Panelist) Uta Gerhardt, (Panelist) Ida Harper Simpson, (Panelist) Jennifer Platt, (Panelist) Barry V. Johnston, (Panelist) Jill M. Niebrugge-Brantley, (Panelist) Patricia Madoo Lengermann
The celebration of the 100th anniversary of the American Sociological Association in 2005 provided the occasion for a multitude of sessions and research projects on various aspects of the intersection of history and sociology. The 2006 meeting builds on this by holding a didactic seminar on the methodologies of the history of sociology, putting in relief how established scholars have undertaken historical research at various levels of the sociological enterprise.
After an initial introduction to linkages between history and sociology, presentations will be made at three levels of historical investigation. Intellectual biographies (respectively those of Sorokin and Parsons) will provide materials for micro level investigations. Meso level methodologies will be illustrated in studying the history of a regional professional association, on the one hand, and studying the history of a social group (women sociologists), on the other. Macro level methodologies will be discussed in researching comparative historical materials with different foci, such as revolutions or methodological issues in the development of an international sociology. The levels and the methodologies deployed are intended to be heuristic for other possible referents, for example, the history of sociological departments and sociological schools.
Among those making presentations will be Charles Tilly, Craig Calhoun, Jack Goldstone, Uta Gerhardt, Ida Simpson, Jennifer Platt, Barry Johnston, Jill Niebrugge-Brantley, and Patricia Lengerman.
