TER General Board
i'm SO relieved that i'm not the only one with a logically precise and razor-sharp protocol when it comes to using the database. it's anal-retentivity of such high order
some people asked me (a) "how did you do that?" and (b) "how long did it take?" answer: using the tools at my disposal and about 10 minutes each.
give a man a new tool and before a woman has put down its instruction manual, he has not only discovered how it works (without ever reading the manual) but he's already figured out ten other ways it can be used. now make that tool have something to do about sex and getting laid ... the you can just forget about it. he will use it nine ways to sunday for every waking second of his horny life! LOL
--
ladies, please do not foam at the mouth over this 
it's a simple good-natured statement of fact. that's why we have tool sheds and an endless array of gadgets in the garage and you must seek our help in programming your VCRs or installing a new OS! 
following up on the post below on "Mathematics of Beauty: Lesson #1" and comments by a few individuals (thank you) i took it upon myself to compute the %s of the ten categories of numerical appearance ("beauty") scores. it's mostly for amusement and/or enlightenment, perhaps also for discussion (that's why i thought it deserved its own separate posting). in keeping with the time-honored academic traditions that a fool like me is about to mimic, i will impartially refrain from any further comments or conclusions to be drawn from this data - at least in this post! LOL. who knows, i may do one of these for performance a bit later ... and before you ask, the answer is YES! this HAS been a slow day at work -- how could you tell?
Graphical Plot *
-----------------
1
1
2
2
33
33
44444
44444
55555555555555555
55555555555555555
666666666666666666666666666666666666
666666666666666666666666666666666666
77777777777777777777777777777777777777777777777777777777777777777777777
77777777777777777777777777777777777777777777777777777777777777777777777
88888888888888888888888888888888888888888888888
88888888888888888888888888888888888888888888888
99999999999999999
99999999999999999
XX
XX
(*) Notes: bar length (in chars) is roughly twice the % value. X stands for 10
IMPORTANT: the samples in this study were the # of PROVIDERS that had *consistent* scores of the respective categories. in other words, the total count for a 7 reflects the number of PROVIDERS (not reviews) that had a consistent ALL-7 rating in their profile (see the Addendum posted in this thread)
Raw Data
---------
Total of 11,681 *providers* from the ten score categories (searched with both exclude options set to NO). note that this data is current as of this day only (see time stamp in the header)
Raw Counts & Percentages:
1 - 30 ~ 0.26 %
2 - 56 ~ 0.48 %
3 - 136 ~ 1.16 %
4 - 319 ~ 2.73 %
5 - 1020 ~ 8.73 %
6 - 2119 ~ 18.14 %
7 - 4136 ~ 35.41 %
8 - 2765 ~ 23.67 %
9 - 981 ~ 8.40 %
X - 119 ~ 1.02 %
-- Modified on 7/16/2003 8:04:22 AM
i'm SO relieved that i'm not the only one with a logically precise and razor-sharp protocol when it comes to using the database. it's anal-retentivity of such high order
some people asked me (a) "how did you do that?" and (b) "how long did it take?" answer: using the tools at my disposal and about 10 minutes each.
give a man a new tool and before a woman has put down its instruction manual, he has not only discovered how it works (without ever reading the manual) but he's already figured out ten other ways it can be used. now make that tool have something to do about sex and getting laid ... the you can just forget about it. he will use it nine ways to sunday for every waking second of his horny life! LOL
--
ladies, please do not foam at the mouth over this it's a simple good-natured statement of fact. that's why we have tool sheds and an endless array of gadgets in the garage and you must seek our help in programming your VCRs or installing a new OS!
i'm surprised no one (not even Staff) has yet asked why with over 100,000 reviews in the database, this analysis was based only on 11,000+ reviews. not to mention that the total number of samples (for appearance and performance are off by a handful - shouldn't the sample sizes be the same?)
well, perhaps i should clarify. while i can not quite answer the discrepancy in the sample sizes (for appearance/performance) i can answer the main question:
(****) the samples in this study were the # of PROVIDERS (and NOT reviews) that had *consistent* scores of the respective categories. in other words, the total count for a 7 reflects the number of PROVIDERS (not reviews) that had a consistent ALL-7 ratings in their profile. the analysis by total # of reviews (regardless of providers) would require access to that oh-so-precious database that TER owns and copyrights, which obviously i don't have (don't worry, just give me time -- no just kidding, please don't cancel my membership??? LOL
---
DISCLAIMER: these statistics show the # of PROVIDERS (not total # of reviews) for each score category. furthermore, these were providers that had CONSISTENT scores (all 7s, or all 2s, etc.)
DICUSSION: what does this mean? strictly speaking it's not a direct reflection on how guys give ratings (though it should be correlated indirectly). for that we would need to look at all (reportedly) 100,000+ reviews (and only TER can do that). however, taken as a random sample, this analysis should still be somewhat representative of the ratings structure overall (with some oblique caveats that i think only i care about
FOR MATH-GEEKS: the situation is somewhat muddled since the number of consistent ratings is indicative of how "popular" or "pernicous" a provider is (strangely enough). naturally, the uncertainty (error bars) on each provider's rating (beyond the YMMV factor itself) is inversely proportional to the square root of the number of consistent reviews.