IIRC, the marriage literature found that opposites do not attract, so you could just give everyone a Big Five and try to minimize distance. As far as seating goes, I'd guess that Extraversion is the most important factor, so put all the extroverts together and the introverts together?
I recall seeing opposite advice regarding extraverts and introverts on some TV show about dinner parties a long time ago. They suggested alternating them when arranging a table - the extroverts should be close enough to talk, but they have to talk around introverts. Introverts would have the opportunity to fluidly join conversations going on around them and have an easier time disengaging because someone else beside them will fill in the gap. Obviously, this would be miserable for an extreme introverts, but you can put those people in a corner.
Again, this is a half remembered piece of information, but it stuck with me because it seems to make sense.
I agree - when I'm in a group of quiet people (regardless of whether they are technically introverts or not), it tends to be a lot more fun if there's at least one talkative person around.
edit: Though I guess things like quiet/talkative are more fluid than introverted/extroverted and partly depend on whether the people in question have anything in common (that's common knowledge between them).
Shoot from the hip guesses: Sort by openness and conscientiousness in a like-with-like way for compatibility. Sanguine agreeables go in the hubs and high traffic areas so they can brighten people's day while neurotics and disagreeables go to different corners and edges where they will have less connectivity to bother or be bothered by anyone other than people with similar problems. I had the same thought as Cog on trying to mix the extraverts and introverts so the outgoing people can draw out the shy ones sitting next to them.
Validated test items for these and many other dimensional breakdowns can be found in the International Personality Item Pool. If anyone is feeling creative (Like to solve complex problems, Ask questions that nobody else does, Know the answers to many questions, Challenge others' points of view, Can easily link facts together), you might try try exploring the index of scales to see if any jump out :-)
Divide the groups in two based on familial affiliation (they'll expect that).
Ask the following questions:
Bias x by (1) [near aisle should be "tiny"] and y by (2) [back should be "tall"]. Average groups, ignore children.
You could check out YourMorals. It's a collaboration among social psychologists who study morality and politics and has lots of tests.
I like that site a lot, but I don't think it will be too useful to alexflint for this.
Here's my current proposed list of statements for people to (strongly) agree/disagree with -- no neutrality allowed. The idea is to identify people on a farmer/forager axis, and then to cluster people according to similarity (subject to other constraints).
Is the farmer/forager distinction very well supported? (And I don't mean some random musings on Robin's blog.) And have these statements been validated for the distinction?
AFAIK, no to the former, and definitely no to the latter: I just made them up. I was hoping that the statements would provide interesting discussion points in addition to personality discrimination. Big 5 tests seem like they might be a little bit less fun for guests to fill out. I acknowledge my lack of academic rigour here.
I was hoping that the statements would provide interesting discussion points in addition to personality discrimination.
Probably will the former, but I'd absolutely get rid of #2 and #4 - politics is the mindkiller, and it's no accident that etiquette manuals proscribe politics during polite discussions.
thanks gwern. I have replaced #2 and #4 with I would like more nuclear power in my country. and I keep the place where I live clean and orderly. Thoughts? This probably needs to go live on the weekend.
My understanding is that there are no standardized personality tests with strong evidence of validity. Using a standardized test might not have any statistical advantages over your own judgement.
ADDENDUM:
If you're just looking to avoid fights, ask people to note down their top ten bands, politicians, and books, and solve for minimum neighbor distance. Then put all the single people together and all it a day.
The wedding in question is actually mine; I'm very grateful to alexflint for posing the question here.
The plan is to provide a brief online (fun, hopefully) test to all guests, and then use a combinatorial optimisation algorithm to assign seating so as to maximise conversational compatibility.
I didn't see any google results for conversation compatibility, so I suspect that you could just make your own - it doesn't have to be perfect at all. Possible categories of conversations you could encourage could be things like books, music, food, relation to the new bride and groom (congrats mosb), human psychology, or travel.
Does anyone know of a freely available, short personality test that would be appropriate for estimating pairwise compatibility for wedding seating?