Re: [Eligibility-discuss] [Gendispatch] New Version Notification for draft-knodel-nomcom-gender-representation-00.txt

[Brian E Carpenter <[e-mail]>]

On 28-Dec-23 02:45, Stephen Farrell wrote:
> 
> Hi SM,
> 
> On 27/12/2023 11:13, S Moonesamy wrote:
>> I read a paper [1] written by Baron, Ganglmair, Persico, Simcoe and
>> Tarantino.  One of the findings in the paper was that female
>> representation in the Nomination Committee causes a decrease in female
>> appointments before 2012 and an increase thereafter.
>>
>> There was a significant event around 2012 which might have influenced a
>> shift.  There was also a letter [2] after 2012 which put the IESG, IAB,
>> IAB and ISOC on notice.
> 
> Thanks for the reference. That's an interesting read that I'd
> not seen before. I'd recommend reading it for others who've not
> seen it and who are contributing to this discussion.

I have substantial doubt whether the conclusion of that paper
has scientific merit. Certainly there is an apparent inflection
point in the statistics that coincides with what the authors call
"the 2012 norm shift" but I see no evidence that this is due to a
causal relationship rather than coincidence. Another hypothesis
is that the small change in the NomCom results since 2013 was a
direct result of the wider societal change shown in Fig. 1, so
nothing to do with the IETF itself and its procedures. A third
hypothesis is that the variation shown in the left hand side
of Fig. 2 is just randomness. The data series is too small to
say.

Generally speaking the whole paper uses "cause" and "effect"
very liberally but doesn't justify them. Of course, that's an
intrinsic problem for many papers in the social sciences. As far
as I can tell, this one has not been peer reviewed.

> My take away is that it supports the idea of the kind of small
> tweak Mallory has proposed, and secondly that for others who'd
> like to see progress on other axes of diversity, maybe arguing
> for the kind of social norm changes within the IETF that are
> discussed in the paper is the way to go, rather than try make
> process change proposals the place from which to start.

Certainly, nothing will change without changes to the social
norms. And I still support Mallory's proposal.

    Brian