[open-science] real consequences of closed science

Mon Sep 11 09:55:32 UTC 2017

Hi everyone,

I've mostly just lurked on this mailing list, but I wanted to share with
this community an article written yesterday by Wired Magazine, about my
home at MIT, a dormitory named Senior House, and how its shut-down reflects
a greater trend for removing counter-culture elements at American
universities: https://www.wired.com/story/a-weird-mit-dorm-dies-and-a-
crisis-blooms-at-colleges/

There are lots of factors, and lots of politics, at play here, but in
essence, the oldest dormitory at MIT (built in 1916) was shut down by
administrators who claimed that rigorous data backed their decisions, yet
they refuse to share their data or their analytical methodology to allow
for independent validation of their conclusions. This is super troubling,
especially given that the MIT administration both seems to have violated
research ethics, and is refusing to acknowledge data that doesn't support
their conclusions.

excerpt from the article (emphasis mine):

In her original correspondence with the Senior House community, and in
> subsequent in-person meetings with the house, [MIT Chancellor Cynthia]
> Barnhart emphasized that the conclusion that Senior House needed a
> turnaround was based on rigorous data—both the graduation rate data from
> the registrar’s office and a campus-wide survey students had taken in 2015.
>
> The questionnaire, the Healthy Minds Survey, was administered by the
> University of Michigan. Many schools around the country give it to students
> as a way to pinpoint problems on campus and decide how best to allocate
> resources. When MIT administered it in 2015, they told students that it was
> a confidential survey intended to help them. One of the chancellor’s
> assistants who had lived in Senior House when she was an undergraduate went
> to Senior House and specifically requested that the residents take it. They
> did, in large numbers.
>
> *What they didn’t know—and what they couldn’t have known from reading the
> consent form that accompanied it—was that MIT had embedded metadata* that
> allowed the administration to pinpoint the location of those filling out
> the questionnaire, enabling them to segment the results by dorm. The only
> question about dorm type in the survey was vague—“What kind of dorm do you
> live in? Small, large, off campus?”—but by tracking the metadata, Barnhart
> and the administration were able to see exactly where respondents lived.
>
> It was this data that enabled Barnhart to see what she called a troubling
> hot spot of drug use. “If it wasn’t a direct violation, it was at least a
> violation of the spirit of informed consent,” Johnson says.
>
> Senior House defenders tried to use this issue to attack the
> administration’s closure of the dorm. “I am extremely angry that MIT used
> data obtained in a questionable manner to inform their policies with
> regards to any of their students, regardless of their residence,” wrote MIT
> parent Elizabeth Glaser in an op-ed in The Tech. Though her son was not
> part of Senior House, he took the survey and was upset about how it had
> been used. Glaser, herself an expert in research ethics, took it upon
> herself to contact the creators of the survey and review its methodology.
> It was she who discovered the metadata.
>
> Critics of the administration also took issue with the data purportedly
> showing Senior House had a relatively low graduation rate. Some worry it
> was based only on where students lived their freshman year, not taking into
> account that some people do switch dorms. Barnhart says the data accounted
> for this.* When WIRED asked for access to the data to analyze the
> methods, the administration declined.* More troubling to critics is that,
> based on the way the data was presented to the student body, it doesn’t
> appear to take into account that the students in Senior House tended to be
> marginalized in one way or another—and that those students tend to have a
> lower graduation rate. Barnhart says that the school took that into
> account as well, looking at marginalized groups in other houses. Again, *it’s
> hard to judge who is correct without access to the raw data or detailed
> information about who and how it was analyzed. *
>

This sort of closed-door, back-rooms strategy doesn't just damage
scientific understanding or career trajectory -- in this case, it also
kicks people out of their homes and endangers their well-being. If you've
ever wondered why I get emotionally worked up about open science, this
whole situation with Senior House is just one example of how closed science
has personally fucked me up and fucked over a community I care deeply
about.

******
*TL;DR -- *closed science has very real and destructive consequences, and
makes it easier for scientific methods and data to be abused and used for
oppression. Claims made in the name of science have great power; thus,
those of us empowered to use science carry a great responsibility towards
the rest of society.

For me the core essence of open science is "*the expectation that a
scientific conclusion is not valid until it can be replicated by an
independent third-party, using either the same dataset and analysis, or a
dataset and analysis that matches the original's parameters*". Given that,
open science can work as a check and a balance against the abuse of
scientific claims. Looking at the current state of politics, in both
science itself and in the greater geographical context, this check is
sorely needed.

Happy Monday,
Danbee

-- 
Brain and Cognitive Sciences, S.B. | MIT Class of 2009 | danbee at alum.mit.edu
Neuroscience PhD candidate | Champalimaud Centre for the Unknown |
Sainsbury-Wellcome Centre for Neural Circuits and Behavior | Intelligent
Systems Lab <http://www.kampff-lab.org/>
단비 | m: +1 (330) 828-6268 | danbeekim.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20170911/f365e7b4/attachment-0002.html>