Researchers have shown increasing interest in technology use among youths and emerging adults (Bleakley, Merzel, VanDevanter, & Messeri, 2004; Redpath et al., 2006; Valentine & Bemhisel, 2008). However, no research to date examines this issue among emerging adult homeless. As these young people are more difficult to engage in services (Ensign & Bell, 2004; Hudson, Nyamathi, & Sweat, 2008) and have higher rates of mental illness (Merscham, Van Leeuwen, & McGuire, 2009) and substance abuse (Barczyk & Thompson, 2008) than the general population, using technology may provide a novel means to approach them. This exploratory study sought to answer two questions: (1) How often, where, and for what purpose do emerging adult homeless use technology? (2) What risk factors (for example, transience, mental illness, addiction) predict technology use? METHOD Sample and Recruitment Using purposive sampling, we recruited 100 emerging adult homeless from Denver (n = 50) and Los Angeles (n = 50) from shelters, drop-in centers, and street outreach (Bender, Ferguson, Thompson, Komlo, & Pollio, 2010; Ferguson, Bender, Thompson, Xie, & Pollio, 2011; Ferguson, Jun, Bender, Thompson, & Pollio, 2010; Sheehan et al., 1998; Thompson, Jun, Bender, Ferguson, & Pollio, 2010). To meet inclusion criteria, participants needed to be in the age range of 18 to 24 years, have spent at least two weeks away from home in the previous month, and provide written informed consent. Wherever possible, agency case managers made the determination whether a particular individual was eligible for recruitment on the basis of personal knowledge of the individual and the individual's current state of sobriety. In cases of key informant referral, the interviewer made the determination. Data Collection and Measures Research staff administered a 45- to 90-minute retrospective interview with participants, who were compensated $10. Both study design and data collection have been described elsewhere (Bender et al., 2010). Human subjects approval was granted by each researcher's university. Technology use was measured by four items (number of days per week you use e-mail, the Internet, a computer, or MySpace). Because nearly half the sample reported daily use on at least one variable, rather than treating the variable as interval-continuous, two dichotomous variables were created: (1) daily use of any technology (0 = everyday, 1 = less than daily) and (2) weekly use of any technology (0 = 1 to 6 days a week, 1 = no days a week). We created the first variable to explore differences between daily users versus all others; the second to compare between regular users (at least weekly) and those not regularly using computers. Three open--ended questions queried the following: (1) With whom do you communicate online? (2) What is the purpose of your online use? (3) Where do you access technology? Predictors of technology use included age, location (0 = Los Angeles, 1 = Denver), gender (0 = female, 1 = male), race-ethnicity (0 = white, 1 = black, 2=Latino), education (0=high school dropout, 1 = graduate or GED holder), current housing status (0 = in stable housing, 1 = homeless or in shelter), transience status (0 = no moves, 1 = at least one intercity move). Data on self reported criminal behaviors (0 = never arrested, 1 = at least one arrest) and social support (frequency of contact: 0 = almost never or occasionally, 1 = often or a lot) were also collected. Using the Mini International Neuropsychiatric Interview (Sheehan et al., 1998), we assessed symptom criteria for posttraumatic stress disorder, mania, depression, and alcohol and drug abuse or dependence (0 = does not meet criteria, l = meets criteria). Data Analysis Independent samples t tests and chi-square or Fisher's exact tests were used to identify differences in characteristics of daily versus nondaily and at-least-weekly versus less-than-weekly technology users. …