Working from home is regaining its popularity because of the advantages it presents for both employees and employers. Telecommunications technologies are enabling the new work-at-home phenomena. This study expands the existing body of work-at-home and telecommuting research by using data from the 1995 Nationwide Personal Transportation Survey to consider a larger sample and to include characteristics unavailable in previous analyses. The effects of socioeconomic, household, locational, and accessibility variables on individuals' choices to work from home are estimated with ordered logit, ordered probit, and multinomial logit models, using a two-equation sample selection regression process. The three models give very similar results. They indicate that educational attainment and the presence of small children in the household encourage frequent working from home. Males and drivers choose to work from home more often than females or nondrivers, and the lack of free parking at work promotes home work. These findings bear implications for trip-generation forecasting and suggest directions for policies intended to influence commute travel.