After decades of stability, the technologies used by workers to locate new jobs began to change rapidly with the diffusion of internet access in the late 1990’s. Which types of persons incorporated the internet into their job search strategy, and did searching for work on line help these workers find new jobs faster? We address these questions using measures of internet job search derived from the December 1998 and August 2000 CPS Computer and Internet Supplements, matched with job search outcomes from subsequent CPS files. We find that internet searchers are positively selected on observables, but negatively selected on unobservables. A beneficial (unemployment-duration reducing) causal effect of internet job search is consistent with our estimates only if negative selection on unobservables is especially strong, in other words only if the population of on-line resumes is strongly adversely selected.