While the analysis may feel about right for the current state of online discussion, I'd like to see the data and the methodology that lead to the analysis so that I better understand why the real names portion, which is +51%/-9%, is colored red and the anonymous portion, which +34%/-11% is colored orange. It seems to me they are backwards. The ranking I'd give using just the few numbers they show here is: Pseudonym use: 61-11=50 (1st rank), Named: 51-9=42 (2nd rank), Anonymous: 34-11=23 (3rd rank). That's be about 43.5% 36.5% and 20% of reputation or lack there of signaling quality. Alternatively you could rank just positive percentages, or just negative percentages coming out with P=41.7%, N=34.9%, A=23.3% or N=29.0%, A=35.5%, P=35.5% respectively where the latter rank is reversed due to being a rank of negative responses. Altogether, the measure of quality itself may not correlate well because there's a possibility pseudonymity, anonymity and use of a real name affects the likelyhood that others are willing to express their positive or negative feelings about the comment, thereby not expressing their feeling about each type of comment evenly.