Posted by Dr. Bob on January 22, 2003, at 19:29:48
In reply to Re: How do people decide which groups to join?, posted by Mitchell on January 22, 2003, at 0:29:39
> > So that would suggest looking at how much conflict and caring there is. Which would be really interesting. But hard to assess?
>
> Yes, at least quantitatively.Hmm, the idea of screening for negative words has come up, that might be a measure of conflict. And I suppose an analogous list of positive words could be used to try to quantify caring...
> Really, the more troubling concern for me is whether conflict from the board is contagious - if excitement started here might play out elsewhere.
>
> Could be the other way too, though. Maybe people have conflicts here instead of with people they see f2f.And both of those could happen with caring as well as conflict...
> > Efficacy is definitely the $24,000 question. And is starting to be looked at. How would you compare administrative policies?
>
> High bid for that answer on Ebay is currently $64,000.I realized later I had underbid, hmm...
> I'm not well read into administrative approaches, but I think one main division would be between styles that ask, require, cajole or manipulate participants to control content and styles in which administrators control content by using word filters, removing posts and such methods.
Yes, but I think it's unlikely to be either-or. And people can be asked as a group (for example, in a FAQ) as well as individually in posts. I guess, as you suggest, those posts could be counted... But not those that have been removed...
Also, administrative activity would depend on not only the site's administrative policies, but also its members...
> You might have already considered measures like ... how many administrative hours compared to group activity.
And if you make it hours, then another factor is how efficient the administrators are...
> > Microsoft Word does tell you the % of sentences that are passive...
>
> I'm not familiar with anything other than standard off-the-shelf grammar analysis tools like that, but I suspect I would find more advanced automated tools for analyzing language if I looked around some.Well, if you do, you could propose a project... :-)
Bob
poster:Dr. Bob
thread:8891
URL: http://www.dr-bob.org/babble/admin/20021128/msgs/8961.html