Exploring day-to-day variability in time use for household members Hejun Kang a,1 , Darren M. Scott b, * a Department of Geography, University of Idaho, Moscow, ID 83844-3021, United States b TransLAB (Transportation Research Lab), School of Geography and Earth Sciences, McMaster University, Hamilton, Ontario, Canada L8S 4K1 article info Article history: Received 3 February 2009 Received in revised form 14 April 2010 Accepted 24 April 2010 Keywords: Activity analysis Intra-household interactions Joint activities Time use Toronto Travel-Activity Panel Survey Variability analysis abstract Studies of activity-travel patterns typically use 1-day or pooled samples, and more often than not, are conducted at the individual level. By default, they assume that activity-travel decisions are uniform from 1 day to the next and individuals are independent from one another. Such assumptions do not reflect reality. This research investigates day-to-day var- iability in activity time-use patterns of household members while incorporating variations in their interactions. Results from a descriptive analysis and a series of daily structural equation models provide evidence of day-to-day variability in activity time-use patterns. Specifically, time-use patterns on weekdays are substantially different from those on weekends. Furthermore, compared to independent activities, there is a higher proportion of intra-personal variability and a lower proportion of inter-personal variability for joint activities. These findings suggest that transportation planners should not combine inde- pendent and joint activities as has been the case in the recent past, nor should they use sin- gle-day or pooled models when estimating activity time-use patterns. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction In travel behavior research, variability analysis, which investigates the extent to which our activity-travel decisions give rise to consistent activity-travel patterns, has been a topic of interest for quite some time (Pas, 1987). The total variability in daily behavior is decomposed into two constituent parts: intra-personal (within-person) and inter-personal (between-per- son) variability (Koppelman and Pas, 1984; Pas, 1987; Pas and Sundar, 1995). Specifically, intra-personal variability refers to differences in the activity-travel behavior of the same individual over time, and inter-personal variability refers to differ- ences in the activity-travel behavior of different individuals on the same day or over different days. Variability analysis, aiming to investigate how activities and travel are organized over a multi-day period, has many advantages (Bhat et al., 2004, 2005; Hirsh et al., 1986; Jones and Clark, 1988; Kitamura, 1988; Koppelman and Pas, 1984; Muthyalagari et al., 2001; Pas, 1986). First, from a policy viewpoint, variability analysis may better reflect changes in the behavioral patterns of individuals in response to policy actions (e.g., work-week compression) than a single-day study (Bhat et al., 2004, 2005; Hirsh et al., 1986) by capturing associations among activities across days of the week. Second, variability studies, which recognize the constraints confronting individuals and households, mediated by the urban area, can help ad- dress the most fundamental issue of travel behavior analysis: why do individuals make trips the way they do? (Kitamura, 1988). Last, it has been found that in travel demand analysis, a multi-day sample, as opposed to a 1-day sample, reduces survey costs and produces more efficient (Pas, 1986, 1987) and less biased estimators (Bhat et al., 2004, 2005; Hirsh et al., 1986). 0965-8564/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.tra.2010.04.002 * Corresponding author. Tel.: +1 905 525 9140x24953; fax: +1 905 546 0463. E-mail addresses: hejun@uidaho.edu (H. Kang), scottdm@mcmaster.ca (D.M. Scott). 1 Tel.: +1 208 885 6452; fax: +1 208 885 2855. Transportation Research Part A 44 (2010) 609–619 Contents lists available at ScienceDirect Transportation Research Part A journal homepage: www.elsevier.com/locate/tra