Formulating friendship

laissez-faire

laissez-faire

laissez-faire

Soldato
Joined
23 Oct 2002
Posts
2,562
Location
Edinburgh/Southampton
Okay - as much as I'd love to say I was doing this because I was some kind of freak trying to rank his friends, I'm currently doing an Economics dissertation.

In order to create interaction models (how playing games with friends affects interactions compared to playing with strangers) I need to 'formulate' the definition of a friend, and be able to define mathematically how close people are.

So, variables to regress : Length of friendship, frequency of interaction, average duration of interaction, average number of people present during interaction, number of common interests.

Throws out some interesting results - that even if you know someone for 5 years but only see them once a month you'll still be 'closer' to them than someone you see every day you have only known for a year. Length you have known someone seems to be the massively significant variable in all the interactions....

Ideas?
 
You cant define mathematically how close people are.

Everyone is different and thus different groups with identical variables will give different results.
 
SiHH said:
You cant define mathematically how close people are.

Everyone is different and thus different groups with identical variables will give different results.
Click to expand...

Well, for economics models, you have to approximate. Let's not get into arguments about whether or not social economics is possible, the fact is that the models are the best way of dealing with many issues in society.
 
Cultural variation is a big thing to look out for here. The Western definition of 'friendship' is probably going to be very different from the Eastern one!
 
Completely and utterly irrelevent. Are you saying I'm closer to a friend I've known for 10 years but only see a few times a year now than I am to my girlfriend of 2.5 years?

How daft.
 
Economics is going to give a complete "pie in the sky" answer.

It's totally pointless.
 
not really up with modelling, but surely it depends on how you weight each of your parameters or are they all given equal weight?
 
An economist knows the price of everything and the value of nothing! You can't use economic theories and models for many real world problems. Economists are a bit like mathematicians- the results work great on paper, but real life isn't like that.

"The Undercover Economist" is a book that highlights these flaws perfectly. The author makes all sorts of arguments that look good on paper but would have big consequences if they were implemented.
 
[TW]Fox;13574904 said:
Completely and utterly irrelevent. Are you saying I'm closer to a friend I've known for 10 years but only see a few times a year now than I am to my girlfriend of 2.5 years?

How daft.
Click to expand...

No I'm not saying that at all, I'm saying that of all the variables, length of time you have known the person is the most statistically significant. Running a regression on all those factors has led me to reject the null hypothesis on them all, hence their inclusion in the model.
 
Ideally, you need to quantify the transfer of social capital in as much detail as possible.

I'm not sure if number of common interests necessarily has the positive impact on a friendship that you might expect. People like variation but also things to talk about. I think that needs to be more closely considered.

I think also you need to quantify how much intimate time is spent. If you share intimate details you are much likelier to have a strong bond as you're more likely to be able to go back to that person for emotional support, unlike someone who you aren't that close with and just have lots of group social interactions.
 
You must log in or register to reply here.
Back
Top Bottom