Syllabus

Sequence Analysis Workshop

Silke Aisenbrey & Anette Eva Fasang

This workshop introduces sequence analysis for social science research. Sequence analysis, originally developed in biology to analyze strings of DNA, has attracted increasing attention in the social sciences for the analysis of longitudinal data. Most applications in the social sciences study life course processes, such as labor market careers, educational careers, or family formation. This workshop covers basic techniques of sequence analysis as well as recent methodological developments tailored at social science research questions. The workshop introduces techniques for sequence comparison, including optimal matching analysis, dynamic hamming matching, sequence similarity based on subsequence metrics, and multichannel sequence analysis. We will also discuss measures for sequence complexity over time, bootstrapping of sequence distances, and sequence visualization. All methods are demonstrated with hands-on examples using R’s TraMineR package.

Version of this workshop have been taught at:

Center for Research on Inequalities and the Life Course, Yale University, New Haven.

Robert Wood Johnson Foundation Health & Society Scholars Program, Columbia University, New York.

Turku Center for Welfare Research, Summer Workshop 2011: "Life course dynamics and the mechanisms of social inequality". University of Turku, Finland.

Day 1

Session 1: morning (non-computer lab)

From Dance Patterns to Career Patterns

- Concepts of time in longitudinal data analysis

- Introduction to optimal matching analysis

- Discussion of different examples of applications include: patterns of lynching, retirement, chronic mental illness, dance, musicians, employment careers, welfare states, family formation

Session 2: afternoon (computer lab)

Hands on Example: Optimal Matching Analysis

- Introduction to R

- Describing and visualizing sequences

- An example application of Optimal Matching Analysis

- We will use a dataset on labor market entry processes for this part of the class.

Day 2

Session 3: morning (non-computer lab)

The ‘Second Wave’ of Sequence Analysis: recent methodological developments tailored at social science research questions

- Time dependent cost specification/dynamic hamming matching

- Sequence similarity based on subsequence metrics

- Multichannel sequence analysis

- Bootstrap Confidence Intervals for sequence distances

Session 4: afternoon (computer lab)

Hands on Example: Above and Beyond Optimal Matching Analysis

- We will use this last session to apply all or some of these more complex sequence methods to an example dataset and possibly data that participants will bring to class (if you have a dataset and are interested in using it in class, please contact the instructors beforehand)