Telephone remains a dominant survey mode for surveys of the general public since most households have a telephone; only 4.8% of households nationwide do not have a phone (US Census).  Some households have only a landline telephone, some have only a cell phone, some have both a landline and a cell phone, and some have multiple landlines and cell phones.  Nationally, about 62% of households have their telephone number listed in a telephone directory (personal communication, Marketing Systems Group).  Cell phone only households have increased from about 4% in 2004 to over 14% in 2007 (Blumberg, et al., 2009).  Given these changes in cell phone only households, should we be concerned about the possibility of coverage error in random digit dialing (RDD) samples?