Nick Holmes 1233218597|%e %b %Y, %H:%M %Z|agohover
The interesting things we could learn from the results would require knowledge of the number of returns and the counts by persona for the bars illustrated. I think the 20% "none of the above" could be important, if it falls entirely within one persona.
We need to know much more about the sample of respondents to judge how representative the results might prove within the user communities / personas.
When considering the outputs, to be able to call down custom ready disclosure controlled tabulations would cut the need to call for so many special tabulations and the associated delays whilst disclosure issues were sorted out - or not in many cases.
There needs to be thought on how to use the provision in the Statistics and Registration Services Act with regards approved researchers (section 39 (4) i). This status allied with an approval of the setting in which the data would be processed, could provide the reassurance to enable less disclosure controlled (sensitive) data sets to be available to researchers not in ONS control centres?
Nick Holmes, 1233218597|%e %b %Y, %H:%M %Z|agohover
