the semistructured section includes questions about how and why information and support has been accessed. Interviews are transcribed professionally, checked by the researcher, and returned to the respondent for change or approval before analysis. For this study, we used DIPEx interviews with respondents who had had cancer diagnosed after 1992. We excluded respondents with cancer diagnosed before 1992 because internet use was then relatively unusual. The interviews included in this study were collected between January 2001 and November 2002. The table shows the distribution and age range of respondents. We analysed interviews for five different cancer groups: men with cancer of the prostate or testis, women with breast or cervical cancer, and 17 men and 14 women with bowel cancer. Consent procedures are approved by Eastern MREC (multicentre research ethics committee), and all interviews are copyrighted to DIPEx for use in research, teaching, and broadcasting. Analysis We read the interview transcripts repeatedly and coded them for analysis. 14 We scrutinised each extract that described use of the internet for any purpose and then coded it under several antici-pated themes (for example, decision making about treatment, information about the cancer) and emergent themes (such as guidance about how to talk to children about cancer, campaign-ing for wider recognition of symptoms). We used constant com-parison in our analysis to ensure that the thematic analysis represented all perspectives. 15 We also sought and discussed negative cases. We did not use significance tests to explore rela-tionships within the data because maximum variation sampling was used for each of the five cancer types we analysed; this sam-pling procedure is designed to include the widest practical range
This is the end of the preview. Sign up
access the rest of the document.