Scott Raney
2016-10-09 16:02:42 UTC
After reading Patrick's message it occurs to me that people may not
really grok what's involved in my proposal, so I invite you to take a
Big 5 inventory to see what it'll actually be like. There are several
of these on the Internet, but they're all basically the same (e.g., my
results are very consistent between them). They all claim to take
about 10 minutes, but if you're any good at taking tests it really
only takes about half that. I recommend:
http://www.outofservice.com/bigfive/
At the end of that you get the raw numbers in the URL line. Mine says:
?oR=0.875&cR=0.833&eR=0.562&aR=0.25&nR=0.281
This in the "OCEAN" order:
Openness: 0.875
Conscientiousness: 0.833
Extroversion: 0.562
Agreeableness: 0.25
Neurosis:0.281
Please show us yours, then tell us:
1) Do you agree with the assessment?
2) Would you be concerned about having this information become public?
As for me, I don't worry about #2 because I know that anyone with even
a little bit of skill in this field could make a very accurate
prediction of these values just from reading my posts to this list.
But not everyone posts (anywhere) and those are exactly the people
(probably low Extraversion and/or high Neuroticism) who are most
poorly represented in our current government systems. One of the
"known unknowns" is how people will react to being asked to do this,
and how extensive the marketing effort that would be required to
convince them that it's really OK (again, it's not anything that
people who have spent more than 5 minutes with you in person don't
already know. Certainly it's far less personally revealing than
posting a picture of yourself on Facebook).
We could potentially save participants some time by cutting out the
questions for 2 of the 5 dimensions (Extroversion and Agreeableness)
which only have a low correlation with ideology, though I'm not as
sure that it won't matter in my direct-voting proposal, which will
almost certainly pick up on things that ideology/party affiliation
measures are simply missing.
There's also a "short form" that sacrifices a little bit of accuracy
to save about half the time (they claim 5 minutes), but again, I'm
inclined to use the standard version to improve accuracy and also
collect the full set of data to make this research publishable
eventually:
http://gosling.psy.utexas.edu/scales-weve-developed/ten-item-personality-measure-tipi/
For further exploration on the subject I recommend
http://www.truity.com/. They also have a Briggs-Meyers test, an older
but more comprehensive inventory/theory. Also see the "Personality
Types" tab and see if you think you can make voting predictions based
on the 16 types. Unfortunately for all the stuff on that site, there
is very little about how personality and politics interact, something
I find generally true of the research in this field. They probably
rightfully reject the idea that choice of politician in a conventional
voting system tells you much about people, but seem to never consider
how personality will make a very big difference in a true direct
democracy.
Note that none of this precludes using different personality
inventories instead of/in addition to big 5. In fact in the long run
I'm sure we'll need to develop a custom inventory that's specifically
designed to facilitate matching on public policy issues. But that's a
big project and would require massive amounts of data to achieve
anywhere near what I believe we can achieve immediately with the
incredibly well-validated big 5 inventory.
Regards,
Scott
really grok what's involved in my proposal, so I invite you to take a
Big 5 inventory to see what it'll actually be like. There are several
of these on the Internet, but they're all basically the same (e.g., my
results are very consistent between them). They all claim to take
about 10 minutes, but if you're any good at taking tests it really
only takes about half that. I recommend:
http://www.outofservice.com/bigfive/
At the end of that you get the raw numbers in the URL line. Mine says:
?oR=0.875&cR=0.833&eR=0.562&aR=0.25&nR=0.281
This in the "OCEAN" order:
Openness: 0.875
Conscientiousness: 0.833
Extroversion: 0.562
Agreeableness: 0.25
Neurosis:0.281
Please show us yours, then tell us:
1) Do you agree with the assessment?
2) Would you be concerned about having this information become public?
As for me, I don't worry about #2 because I know that anyone with even
a little bit of skill in this field could make a very accurate
prediction of these values just from reading my posts to this list.
But not everyone posts (anywhere) and those are exactly the people
(probably low Extraversion and/or high Neuroticism) who are most
poorly represented in our current government systems. One of the
"known unknowns" is how people will react to being asked to do this,
and how extensive the marketing effort that would be required to
convince them that it's really OK (again, it's not anything that
people who have spent more than 5 minutes with you in person don't
already know. Certainly it's far less personally revealing than
posting a picture of yourself on Facebook).
We could potentially save participants some time by cutting out the
questions for 2 of the 5 dimensions (Extroversion and Agreeableness)
which only have a low correlation with ideology, though I'm not as
sure that it won't matter in my direct-voting proposal, which will
almost certainly pick up on things that ideology/party affiliation
measures are simply missing.
There's also a "short form" that sacrifices a little bit of accuracy
to save about half the time (they claim 5 minutes), but again, I'm
inclined to use the standard version to improve accuracy and also
collect the full set of data to make this research publishable
eventually:
http://gosling.psy.utexas.edu/scales-weve-developed/ten-item-personality-measure-tipi/
For further exploration on the subject I recommend
http://www.truity.com/. They also have a Briggs-Meyers test, an older
but more comprehensive inventory/theory. Also see the "Personality
Types" tab and see if you think you can make voting predictions based
on the 16 types. Unfortunately for all the stuff on that site, there
is very little about how personality and politics interact, something
I find generally true of the research in this field. They probably
rightfully reject the idea that choice of politician in a conventional
voting system tells you much about people, but seem to never consider
how personality will make a very big difference in a true direct
democracy.
Note that none of this precludes using different personality
inventories instead of/in addition to big 5. In fact in the long run
I'm sure we'll need to develop a custom inventory that's specifically
designed to facilitate matching on public policy issues. But that's a
big project and would require massive amounts of data to achieve
anywhere near what I believe we can achieve immediately with the
incredibly well-validated big 5 inventory.
Regards,
Scott