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.
cated by the negative coefficient on the age variable. Men are found to prefer to socialize with non family members to family members (either household or nonhousehold). Caucasians are found to have a lower propensity to engage in social activities with nonhousehold family members compared with individuals of other ethnicities. Employed persons have a higher propensity to choose nonhousehold other members as companions for social activities. This is intuitive given that these companions are often co- workers. Students are more likely to social- ize with friends and less likely to undertake joint social activities with household and nonhousehold family members. Married individuals are more likely to undertake social activities with only household members (see the negative coefficients for all other alternatives). Further, the negative coefficients are strongest for the three alter- natives that do not include household members. This indicates that married individuals participate in social activities with their spouses. Finally, the absence of chil- dren in the household favors socializing with nonhousehold members. Companion- Type Model for Passive Leisure Activities The MNL model for the companion- type choice for pas- sive leisure activities is presented in Table 6. The âsoloâ alternative is chosen as the reference category. The âhousehold members onlyâ alternative is not available for individuals in single- person households. All other alternatives are available for all individuals. Results indicate that passive leisure episodes of longer durations are more likely to be pursued jointly than solo. Further, among the joint episodes, shorter- duration activities are more likely to be pursued with non - household other members as companions, as are week- day episodes. Otherwise, weekday episodes are more likely to be solo than joint. Finally, friends are more likely to be companions than household or family mem- bers for weekday passive leisure activities. Younger individuals are more likely to undertake pas- sive leisure activities with nonhousehold members. Men are found to undertake passive leisure activities indepen- dently or with nonhousehold, nonfamily members as companions. Caucasians have a lower propensity to undertake solo activities compared with individuals of other ethnicities. Employed persons have a higher propensity to choose nonhousehold other members as companions for passive leisure activities. This is intuitive given that these com- panions are often co- workers. These persons also prefer independent leisure to joint leisure with non- co- workers as companions. Students are more likely to pursue leisure with friends and colleagues and less likely to do so with nonhousehold family members. Married individuals are found not to prefer pursuing joint leisure with only friends or nonhousehold family members. Solo episodes are favored over joint episodes with nonhousehold, nonfamily companions. However, joint episodes including household members as compan- ions are preferred to solo episodes. Finally, the absence of children in the household favors pursuit of passive leisure with only nonhousehold friends and family. When children are present in the household, household members are the most favored companions for leisure. Companion- Type Model for Active Leisure Activities The MNL model for the companion- type choice for active leisure activities is presented in Table 7. The âsoloâ alternative is chosen as the reference category. The âhousehold members onlyâ alternative is not avail- able for individuals in single- person households. All other alternatives are available for all individuals. Results indicate that active leisure episodes of longer duration are more likely to be pursued jointly than solo. 134 INNOVATIONS IN TRAVEL DEMAND MODELING, VOLUME 2 TABLE 6 Model for Companion-Type Choice for Passive Leisure Activities Household Nonhousehold Nonhousehold Nonhousehold Mixed Members Family Friends Other Composition Beta t-stat Beta t-stat Beta t-stat Beta t-stat Beta t-stat Constant 1.108 7.670 0.194 1.005 0.288 1.767 0.294 2.004 0.007 0.045 Activity episode duration 0.013 23.617 0.011 18.536 0.013 24.497 0.002 4.066 0.014 25.320 Weekday 1.097 14.038 0.890 10.322 0.297 4.220 0.292 4.840 1.165 15.323 Age 0.018 6.315 0.026 10.116 0.013 6.232 0.022 8.847 Male 0.533 7.800 0.415 5.413 0.629 9.778 White 0.360 3.299 0.192 1.792 0.362 4.049 0.141 2.012 0.369 3.696 Employed 0.926 10.598 0.977 10.348 0.499 6.211 0.478 5.952 0.828 9.965 Student 0.342 2.611 0.587 6.372 0.267 3.231 Married 1.730 18.831 0.262 2.881 1.026 12.067 1.228 16.110 No children in household 0.803 10.183 0.477 4.719 0.280 3.457 0.174 2.852 0.676 8.088 Log likelihood (convergence) 15,404.93 Log likelihood (constants only) 18,158.04
