[gnso-rpm-wg] Member questions (Re: Critique of INTA survey)
icann at leap.com
Thu Aug 31 00:42:35 UTC 2017
On Wed, Aug 30, 2017 at 6:32 PM, Mary Wong <mary.wong at icann.org> wrote:
> To assist Lori and the Working Group chairs, staff intends to compile all
> the questions that are sent to this list so that, if they are not addressed
> on the call later, Lori will be able to review them following the call. This
> may be especially helpful if you raise a question about a particular data
> point or a slide from the Nielsen deck that we are not able to cover on the
> call. It will be helpful therefore if, like Paul, any questions you have
> about a specific result can also include a reference to the slide number.
For each survey result (percentage, average, etc.) that INTA is
presenting and purporting to be statistically valid (which I and Kurt
have already questioned the validity of), I request that INTA also
publish the error margins and confidence level attached to each and
every figure (e.g. +/- X, 19 times out of 20, or whatever), and also
how they derived that confidence level.
The standard calculations require that the sample be random, too,
which was not the case in this study (as previously discussed). So
please also point out any adjustments to, if any, that were made to
the standard calculations that back up the figures.
The above obviously can't be answered during the call by Lori, but can
be done sometime after the call.
More information about the gnso-rpm-wg