LOG-LINEAR APPROACHES TO LAG-SEQUENTIAL ANALYSIS WHEN CONSECUTIVE CODES MAY AND CANNOT REPEAT

Authors
Citation
R. Bakeman et V. Quera, LOG-LINEAR APPROACHES TO LAG-SEQUENTIAL ANALYSIS WHEN CONSECUTIVE CODES MAY AND CANNOT REPEAT, Psychological bulletin, 118(2), 1995, pp. 272-284
Citations number
42
Categorie Soggetti
Psychology,Psychology
Journal title
ISSN journal
00332909
Volume
118
Issue
2
Year of publication
1995
Pages
272 - 284
Database
ISI
SICI code
0033-2909(1995)118:2<272:LATLAW>2.0.ZU;2-0
Abstract
G. P. Sackett's (1979) lag-sequential analysis is recast in log-linear terms. Differing from articles by P. D. Allison and J. K. Liker (1982 ) and by D. lacobucci and S. Wasserman (1988), sequences that, for log ical reasons, do not allow consecutive events to be assigned the same code are considered. In addition, overlapped sampling, which the seque ntial analysis literature usually assumes, is contrasted with nonoverl apped sampling of sequences, which is also used and may seem more fait hful to some statistical models. Several advantages of a log-linear ap proach to sequential problems are noted, including its ability to deal routinely with the structural zeros created when consecutive codes ca nnot repeat and its integration of sequential methods into an establis hed and well supported statistical tradition.