Cancer strategy: consultation analysis

Methodology

The Scottish Government's online consultation was hosted on the Citizen Space portal and consisted of 11 closed-format and 33 open-format questions. Responses could be submitted via the online platform or directly to the Scottish Government by e-mail (to a dedicated consultation inbox) or post. Responses from Citizen Space and those sent by e-mail and post were merged into a single, final dataset combining all responses to the consultation.

As is usual for a large-scale government consultation, there were a small number of responses that shared identical or near-identical text, though it is difficult to robustly identify whether these responses came from coordinated campaigns or individuals and/or groups who informally consulted each other before submitting their responses. Because no two respondents submitted close or exact duplicate text across all questions in their responses, this suggests that the results of the consultation reflected the views of a range of organisations and individuals rather than being dominated by specific campaigns. Due to the small sample sizes involved and to ensure that all responses were considered, responses which were close or exact duplicates were only removed from the thematic analysis (and not the quantitative analysis) presented in the report.

Descriptive analysis was conducted on the responses to the 11 closed-format questions using Python. The main body of this report presents a breakdown of responses to each consultation question. Each question includes a chart that summarises responses as a percentage of all consultation respondents who answered the question. Full tables of results are presented in Appendix A.

For the 33 open-format questions, there was no limit to the amount of text in which respondents could write in their answers. To synthesise important themes/perspectives raised across respondents for each question, we followed a thematic analysis approach based on Braun and Clarke (2006) that involved four phases:^[7]

1. Manually reviewing each free-text response to highlight patterns/recurring themes as well as singular cases (ideas/perspectives not raised in other responses).

2. Mapping the qualitative themes to specific components of the consultation and developing a narrative description for each theme.

3. Triangulating themes within/across respondents and assessing their substantive significance based on convergence/divergence of perspectives.

4. Identifying any insightful outlier responses that do not fit in with the general emerging themes and analysing patterns of non-responses (if there are any).

The total number of respondents answering each open-format question is listed at the beginning of each section. This total includes respondents who wrote in "No comment", "Nothing to add", "N/A", etc., though in general these responses made up less than 10% of total responses to open-format questions.

There were 12 consultation responses submitted via e-mail with PDF attachments, and respondents did not indicate which questions of the consultation they were responding to in the PDF attachments. We reviewed these responses in full, and themes were assigned to one or more individual consultation questions where possible. In general, the key points and ideas raised in e-mailed consultation responses did not differ from those raised in responses submitted via Citizen Space.

The main body of this report follows the same question order as the consultation document, and themes for each open-format question are presented in order of frequency (number of responses that discussed the theme). Specific themes are highlighted if the proportion of individuals or organisations mentioning these themes differs from the overall proportion of individuals which answered the question. In the qualitative analysis, the research team has sought to be respectful of the full spectrum of emotions, views and perceptions expressed within the consultation. Written responses were self-selecting, and it is understood that these do not necessarily represent the wider public opinion in Scotland. Individual quotes have been used, where appropriate, to illustrate the narrative around specific themes, and quotes were only selected from respondents who provided permission for their views to be published and with any potential identifiers (such as the name of a specific organisation) removed. Where the quotes selected included typos, these were corrected by the research team to allow readers to read the views shared uninterrupted.