Posted by Estella on August 29, 2006, at 21:40:00
In reply to Re: blocks and support Estella, posted by Jost on August 28, 2006, at 19:40:11
> Critiques include positive as well as negative comments.
Yes indeed. It is hard to motivate someone to look for a better strategy if people aren't aware of what may be wrong with the current strategy
Currently, however, people get blocked from the politics board for trying to point out unsavory aspects of current strategies so it seems that only POSITIVE comments about policies, institutions, and ideologies is permissible. The trouble with that is that we simply aren't allowed to critique on the politics board.
> Also, it's not so simple to design tests to show that things such as blocking people are working, or that such and such empirical test has any necessary or compelling relationship to what you're trying to prove.
I know it isn't so easy, but I figure it would be possible to do a reasonable study (which is to say about as reasonable as lots of other studies out there). If you think there is a causal relationship between blocking and certain things like reoffending, leaving the boards etc then it would seem to me that the place to start would be to have a look at the stats of people reoffending, leaving the boards etc. While correlation doesn't imply causation it would be interesting to see WHAT correlations obtain in the first place. If there are correlations then the next step would be to systematically alter the independent variable (blockings for certain kinds of offences) in order to see whether there are systematic alterations to the dependent variables. If we agree that systematic variation is causation (as many scientists seem to accept it is) then a causal relationship between blockings and reoffending / leaving would be established. It would then be an empirical matter as to how to achieve the best trade-off between the positive and negative consequences of blocking. E.g., if you shorten the block lengths there might be more 'minor' offences, but less people may leave as a consequence of the blocking system.
In all this... I think offenses need to be typed into kinds. I think there are some offenses that most people would agree blockings are appropriate for (when people fairly clearly accuse, attack, judge another etc) and there are other offenses where most people agree blockings are not appropriate (critiquing ideologies and institutions for example). Whether you block or not and the length of time you block for would be the independent variables. I think it would be possible to do this. Offenses are typed already to a certain extent. Typing wouldn't seem to be more arbitrary than oh say the way DSM types. It doesn't need to be an exact science, just needs to be good enough...
The first step would be to type offenses. That is probably best done by looking through the archives and seeing what kinds of offences people are blocked for. It might be hard to come up with an appropriate category of types, but I don't think it is impossible or essentially arbitrary. One way to do it would be to see whether one person can organise blocks into types under certain headers and then see whether once the headers are given whether another poster can organise the blocks into the same headers. If so then there is inter-rater reliability for the types at any rate. There might be lots of funny borderline cases but I would say that we should have a go rather than giving up on a-priori grounds.
I'm interested in whether lengthy blocks for certain types of offences actually harm the boards. What I mean by harm is operationalised and fairly specific:
1) It harms the boards when a person who usually posts supportive messages leaves the board for a lengthy period of time / for good.
2) It harms the boards when a person who usually posts supportive messages starts posting unsupportive / attacking messages.
3) It harms the boards when a person who usually posts supportive messages leaves the boards because another poster was blocked for a lengthy period of time for a borderline offence.I think that lengthy blocks for borderline infractions tends to CAUSE both those results. Hence... Blocking people for lengthy periods of time tends to HARM the boards in the above mentioned ways.
Bob thinks that might well be so but he thinks that blocking people for lengthy periods of time actually prevents further incivilities on the boards.
Only one way to find out and yeah it is indeed an empirical matter...
poster:Estella
thread:670602
URL: http://www.dr-bob.org/babble/admin/20060826/msgs/681332.html