ACSPRI Conferences, RC33 Eighth International Conference on Social Science Methodology

Font Size:  Small  Medium  Large

A Permutation Test for Group Comparisons of Social Science Sequences

Tim Liao, Anette Fasang

Building: Law Building
Room: Breakout 4 - Law Building, Room 106
Date: 2012-07-10 01:30 PM – 03:00 PM
Last modified: 2011-12-12


Sequence analysis has seen many developments and increasingly wide applications in the social sciences. However, no formal way exists for comparing groups of social science sequences to determine whether they are different in a statistically meaningful way. To fill this gap, we propose a permutation test for such comparisons. We view a typical social science sequence such as life-course or employment-history sequences as having certain characteristics—transition to first marriage, first birth, or first job, that contribute some unique information, for example. Therefore, in addition to proposing a permutation test for comparing overall sequence-group differences via sequence-based distances in the form of the Levenshtein distance, we propose to devise a permutation test on statistics that isolate specific aspects of sequences. Examples of such statistics include the frequency of a certain transition and the timing of a certain transition. To illustrate the permutation test, we apply it to data from the German Life History Study (GLHS) about family formation among East and West German women.