Innovations in Travel Demand Modeling, Volume 2: Papers (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

hold members as companions. Joint travel during week- days is more likely to be pursued with only non - household members than joint travel during weekends. This is consistent with the findings for the companion types for out- of- home activity episodes. In summary, the results from Tables 2 and 3 highlight that joint activity–travel constitutes a significant propor- tion of individuals’ overall activity–travel patterns. In the next section, we focus on certain leisure activities (social- izing, passive leisure, and active leisure) for further analysis. However, the summary statistics discussed here suggest that detailed analysis of all other non leisure activity types is also warranted. Analysis of Out- of- Home Leisure Activities This section of the empirical analysis focuses on individ- uals’ companion- type choices for three types of out- of- home leisure activities— socializing (visiting friends, attending a party), passive leisure (attending movies, sports events), and active leisure (participation in sports or exercising). The choices and the sample shares for each of these three activity types are presented in Table 4. Socializing activities are joint by definition and are equally likely to be undertaken with family members, non family members, and with a mixed composition. Pas- sive leisure episodes are most likely to be pursued alone or with nonhousehold other members (often colleagues). Active leisure episodes are most likely to be pursued independently. When undertaken with other individuals, the companions are most likely to be household mem- bers or friends. Finally, the last two alternatives have been combined into a single category for the passive and active leisure episodes. Thus, each model has six alterna- tives in the universal choice set. Companion- Type Model for Socializing Activities The MNL model for the companion- type choice for socializing activities is presented in Table 5. The “house- hold members only” alternative is chosen as the refer- ence category. This alternative and the “household and nonhousehold family members” alternative are not available for individuals in single- person households. All other alternatives are available for all individuals. Empirical results indicate that short- duration episodes are more likely to be undertaken with other non - household members (often colleagues), whereas long- duration episodes are undertaken with a mixed composition of companions involving household and nonhousehold family members and others. Weekday episodes are more likely to be pursued with non - household members. As would be expected, younger individuals are more likely to undertake social activities with friends as indi- 133COMPANIONSHIP FOR LEISURE ACTIVITIES TABLE 4 Sample Shares on Companion Type for the Three Types of Leisure Activities Socializing Passive Leisure Active Leisure Freq. % Freq. % Freq. % Solo NA 2,813 26.11 2,233 40.08 Only household members 624 4.95 1,380 12.81 995 17.86 Only nonhousehold family members 2,049 16.25 854 7.93 197 3.54 Only nonhousehold friends 2,580 20.46 1,655 15.36 902 16.19 Only nonhousehold other 2,305 18.28 2,598 24.12 688 12.35 Both household and nonhousehold family members 2,023 16.05 1,472 13.67 557 10Mixed composition 3,026 24.00 Total 12,607 100.00 10,772 100.00 5,572 100.00 NA = not applicable. TABLE 5 Model for Companion-Type Choice for Socializing Activities Nonhousehold Nonhousehold Nonhousehold Household and Other Mixed Family Friends Other Nonhousehold Family Composition Beta t-stat Beta t-stat Beta t-stat Beta t-stat Beta t-stat Constant 1.559 11.847 1.754 12.666 1.457 11.229 1.563 12.087 1.730 14.473 Activity episode duration 0.004 10.508 0.002 6.434 0.001 5.890 Weekday 0.479 8.293 0.739 13.420 1.205 21.818 Age 0.005 2.796 Male 0.146 2.379 0.516 9.002 0.467 8.023 0.140 2.402 White 0.301 4.702 Employed 0.441 8.475 Student 0.553 7.604 0.345 3.738 Married 1.592 12.336 2.296 17.691 1.917 15.008 0.565 4.234 0.576 4.589 No children in household 1.231 18.161 0.877 12.244 0.495 7.586 0.170 2.996 Log likelihood (convergence) 19,263.24 Log likelihood (constants only) 20,949.43