All the analyses
were conducted using SPSS version 12.0. The data met the assumptions of normality
and lacked outliers. Pairwise deletion was used to manage missing data due to
respondent tendency to not answer questions because of a lack of recall and/or
not having been prescribed certain self-care behaviors. Because the surveys that
address the recollections of a family member’s illness representation and the version
that addresses one’s own illness representations ask the same questions, a
principal axis factor analysis of all items (correct) was conducted to ensure that the
effect sizes were not inflated because of method bias.