Absolute comment values restrict the range of comments, usually placing the dumb first posts and unpopular pro MS comments in the same category as the truly uncivilized comments. My modest proposal to alleviate this problem, along with those of not enough mod points, and points not lasting.
1-Give all logged in users (or the oldest 75% if you wish) the ability to moderate once per comment, either up or down (siskel and ebert style) at all times.
2- tally the net value for each comment (which could run from -1066(c.f Stevens article) to +600(written by Bruce P.) or whatever)
3- Allow readers to set defaults based on percentages rather than absolute intergers. E.g., read the top 50%, the elite 5%, or the whole 100% of comments posted.
Assuming the average/. reader is decent, a relatively normal distribution should result. After a couple of articles have been tested, certain percentage marks will demarcate categories.(e.g., maybe 0-0.05% is usually flaimbait; 5-10% me toos; and >60% insightful and informative).
Then you can track the usage rates of moderation, and see that say the average user moderates 10-20 comments a day (for example). So those who moderate 2 standard deviations or more from this mean can easily be identified and investigated to see if they are misusing moderation (and whup ass accordingly).
idk, just some pre-dinner ramblings. grunt grunt food!
The BBC article references a different study than the yahoo uk article does. The BBC talks about Dr Greenfield's study, which I has the *fortune* to see as he presented. I generally discuss my reaction to both of these in a new thread (which does not appear to be up yet, or i'd show ya the quik way to it)
Coincidently, I just returned from the APA (American Psychological Association) Annual Convention. As a doctoral student in clinical psychology my pet research interest is in psychology of / psychology on the internet, and at the convention I had the oppurtunity to see presentations of the leading American psychological internet researchers.
And they can be just as clueless!
Actually, that is a mild characterization of their research strategy. Stronger characterizations could portray them as irresponsible researchers or economic opportunists.
Here are some of my criticisms of what I have seen / read (to substantiate the previous flammable commentary):
1) use of very large Ns, 9k to 18k, which is very rare in psychological research. With such large Ns, it is very likely that minute numerical differences will be found statistically significant. (So it is more probable to falsely find differences between groups, and for errors to be made).
2) most studies use Survey data, which is NOT scientifically rigorous! Surveys yield correlational data, which does not permit causal inferences. Common in this case: does internet use cause lonliness / depression, does depression cause internet use, or does something else (like lack of a car or geographic isolation) cause both? Survey data can not answer these questions.
3) Many psychologists do not undertand / appreciate the culture of regular or heavy internet users. They discount social contacts formed and sustained via the internet. To most psychologists, real life has no substitute, and internet relations are worthless and misunderstood. (Sociologists seem to be much less clueless and seem to be ahead of psychology at this time).
4) Two of the most prominent studies occurred exclusively on single sites, one on abc.com and one on msnbc.com.
One very real problem of this is selection bias, who do you know that regularly visits these sites? These sites cater to a specific audiece, and that audience may be VERY different from the rest of internet users, thus any results may not apply to people are not in those audiences.
...and, for those inclined to conspiracy theories
I think I remember the msnbc study advertized on the cable tv channel. The researchers do not consider the possiblility of ballot stuffing (by, say, the Religious Right to defeat the porn-riddled internet demon; or by Internet widows, who entered info for their never-available husbands (sorry for the gender stereotyping).
Not to mention net users (especially l33t kiddies) entering false info to screw with the researchers. Or possible effects similar to the/. effect.
These researchers are also not technologically savvy, and they seemed to undertake rudimentary steps to prevent multiple entries / cracking, which may slant the results. (deleting cookies, ip-spoofing, yada yada yada).
Wanna go way out there? NBC and ABC hosted / sponsored these studies, who have economic incentives to see the internet get a black eye.
What about this Belgian study? Having not read the study, I can not speak directly to their research. But I know a little about Dopamine (DA), that makes me skeptical:
1- DA is strongly implicated in subjective feelings of pleasure! (how cocaine works, and maybe sex). So People who are the internet alot are having fun? This must be a problem...
2 - DA deficiency is related to Depression and Schizophrenia. Since researhcers have already claimed the internet causes depression, this coulbe be a contradictory finding.
3 - It is unlikely (probably unethical) that this was an experimental study, where they manipulate internet usage (you must surf the nt for 4 hours now, begin); or DA levels. Thus, this may also be correlational data, mini critique as above #2.
As may be evident, I am very interested in psychological and internet interactions. I am biased in favor of internet based communities, and computer mediated relationships. I am currently working on culturally sensitive internet research, and hope to present at the next Annual APA convention. So place my commentary in the appropriate context.
Slashdot Mirror
My radical 2 cents:
/. reader is decent, a relatively normal distribution should result. After a couple of articles have been tested, certain percentage marks will demarcate categories.(e.g., maybe 0-0.05% is usually flaimbait; 5-10% me toos; and >60% insightful and informative).
Absolute comment values restrict the range of comments, usually placing the dumb first posts and unpopular pro MS comments in the same category as the truly uncivilized comments. My modest proposal to alleviate this problem, along with those of not enough mod points, and points not lasting.
1-Give all logged in users (or the oldest 75% if you wish) the ability to moderate once per comment, either up or down (siskel and ebert style) at all times.
2- tally the net value for each comment (which could run from -1066(c.f Stevens article) to +600(written by Bruce P.) or whatever)
3- Allow readers to set defaults based on percentages rather than absolute intergers. E.g., read the top 50%, the elite 5%, or the whole 100% of comments posted.
Assuming the average
Then you can track the usage rates of moderation, and see that say the average user moderates 10-20 comments a day (for example). So those who moderate 2 standard deviations or more from this mean can easily be identified and investigated to see if they are misusing moderation (and whup ass accordingly).
idk, just some pre-dinner ramblings. grunt grunt food!
The BBC article references a different study than the yahoo uk article does. The BBC talks about Dr Greenfield's study, which I has the *fortune* to see as he presented. I generally discuss my reaction to both of these in a new thread (which does not appear to be up yet, or i'd show ya the quik way to it)
Coincidently, I just returned from the APA (American Psychological Association) Annual Convention. As a doctoral student in clinical psychology my pet research interest is in psychology of / psychology on the internet, and at the convention I had the oppurtunity to see presentations of the leading American psychological internet researchers.
/. effect.
And they can be just as clueless!
Actually, that is a mild characterization of their research strategy. Stronger characterizations could portray them as irresponsible researchers or economic opportunists.
Here are some of my criticisms of what I have seen / read (to substantiate the previous flammable commentary):
1) use of very large Ns, 9k to 18k, which is very rare in psychological research. With such large Ns, it is very likely that minute numerical differences will be found statistically significant. (So it is more probable to falsely find differences between groups, and for errors to be made).
2) most studies use Survey data, which is NOT scientifically rigorous! Surveys yield correlational data, which does not permit causal inferences. Common in this case: does internet use cause lonliness / depression, does depression cause internet use, or does something else (like lack of a car or geographic isolation) cause both? Survey data can not answer these questions.
3) Many psychologists do not undertand / appreciate the culture of regular or heavy internet users. They discount social contacts formed and sustained via the internet. To most psychologists, real life has no substitute, and internet relations are worthless and misunderstood. (Sociologists seem to be much less clueless and seem to be ahead of psychology at this time).
4) Two of the most prominent studies occurred exclusively on single sites, one on abc.com and one on msnbc.com.
One very real problem of this is selection bias, who do you know that regularly visits these sites? These sites cater to a specific audiece, and that audience may be VERY different from the rest of internet users, thus any results may not apply to people are not in those audiences.
...and, for those inclined to conspiracy theories
I think I remember the msnbc study advertized on the cable tv channel. The researchers do not consider the possiblility of ballot stuffing (by, say, the Religious Right to defeat the porn-riddled internet demon; or by Internet widows, who entered info for their never-available husbands (sorry for the gender stereotyping).
Not to mention net users (especially l33t kiddies) entering false info to screw with the researchers. Or possible effects similar to the
These researchers are also not technologically savvy, and they seemed to undertake rudimentary steps to prevent multiple entries / cracking, which may slant the results. (deleting cookies, ip-spoofing, yada yada yada).
Wanna go way out there? NBC and ABC hosted / sponsored these studies, who have economic incentives to see the internet get a black eye.
What about this Belgian study? Having not read the study, I can not speak directly to their research. But I know a little about Dopamine (DA), that makes me skeptical:
1- DA is strongly implicated in subjective feelings of pleasure! (how cocaine works, and maybe sex). So People who are the internet alot are having fun? This must be a problem...
2 - DA deficiency is related to Depression and Schizophrenia. Since researhcers have already claimed the internet causes depression, this coulbe be a contradictory finding.
3 - It is unlikely (probably unethical) that this was an experimental study, where they manipulate internet usage (you must surf the nt for 4 hours now, begin); or DA levels. Thus, this may also be correlational data, mini critique as above #2.
As may be evident, I am very interested in psychological and internet interactions. I am biased in favor of internet based communities, and computer mediated relationships. I am currently working on culturally sensitive internet research, and hope to present at the next Annual APA convention. So place my commentary in the appropriate context.