Questionnaires were distributed as e-mail attachments to per-sonal contacts, and to members of the public on the socialnetworking sites Twitter and Facebook. Other respondents werecollected by snowballing on the basis of the original contact group.Although access was a major driver in the sampling approach, thisconvenience sampling approach was also beneficial in building asample in which respondents all had the shared characteristic ofusing a mobile phone (Riley, Wood, Clark, Wilkie, & Szivas, 2000).Hair, Bush, and Ortinau (2006) suggest that self-selected samplingis suitable for exploratory research when prior knowledge of thepopulation’s characteristics is not sufficiently present.