Last week, the National Center for Health Statistics released new estimates finding that in the second half of 2009, nearly one-quarter of adults (23%) live in households with no landline telephone service. Among some subgroups, the wireless-only numbers are significantly higher—30% of Hispanics and 49% of adults aged 25-29 lack landlines. Pollsters have long used techniques like weighting to compensate for the absence of demographic groups from their samples, but can they still be successful in light of ever-rising number of cell phone-only households?