But using b-live to provide a sample also skews the results: due to the nature of the site, respondents are already engaged with the careers process and have been exposed to the careers that b-live promotes - making them more likely to be interested in those specific roles.
When they logged onto the survey site, respondents first identified the type of organization that employed them.
No significant differences were found between hospital and field site respondents.
The Panzi and field site respondent groups were statistically compared to determine whether significant demographic differences existed between these groups.
The nurse supervisor made the questionnaires available for volunteer nurses to fill out anonymously on site; respondent nurses could ask questions directly to the supervisor, and the supervisors collected even those questionnaires not filled out on site.
Sites and respondents were selected using purposive sampling.
On univariate analysis, in some sites nonadherent respondents were more likely to be African-American, and in other sites they were more likely to be Hispanic.
(M itanin group interview, Durgkondal ) In both case study sites, the respondents emphasized the lack of awareness of and information on rights and entitlements.
