Sorry for the long post here but I thought there would be some interest. I was on the site a couple weeks ago and found the page with the PE data download. I downloaded it and was impressed by the amount of information in the file. Nearly 12,000 records!!
I am not a statistician but I do work with numbers on a daily basis - forecasting, modeling, projecting and basic analytical work. With 12,000 records I could not help myself, I had to try to read into them.
We have all seen the size surveys - average length is X.XX and average girth is X.XX. But I have never seen survey results that combined length AND girth. This data allowed me to do just that.
The first thing I did was remove the duplicates. For members with multiple entries, I used the smallest or earliest entry. This brought the record count down to about 4,900 individuals. From there I removed the really odd entries - 1” girths, 15” lengths etc. There were surprisingly few of those, and I was left with 4,832 records.
The results are bigger than real world norms, but very reasonable considering it is data from dedicated boner builders and most first entries were likely made after some success. The initial results were..
Average Length = 6.4”, Average Girth = 5.1”
Median Length = 6.5”, Median Girth = 5.04
But this does not mean the guy exactly at the 50 percentile mark was 6.5 X 5.04. Average L’s and G’s as normally stated are independent of each other. The database allowed me to analyze L&G together and the actual median size for the group is 6.2 X 4.7.
Only 30% have Length equal to or bigger than 6.5 AND Girth equal to or bigger than 5.04. There were 500 records with lengths ranging from 6.4 to 6.6”. Of those, 36% have girths under 5”, 26% under 4.75” and 11% are over 5.5”.
I created a couple of grids to help me make sense of this (see attachment). One grid is for volume using the (r*r*h*3.1416 formula). I know this does not give 100% accurate volumes but when the same formulas are applied to all dimensions, the results are directionally and comparatively relevant. I also included colored coded percentile ranges in this grid. The other grid is for Length and Girth. Look up your numbers and it will tell you your size percentile. Example: 7 X 5.5 is at the 87% mark - 87% are shorter and/or thinner and just 13% are longer AND thicker.
I threw in couple of graphs laying in measurements from the database compared to the Pre-PE polls. I’d like to do some “adjustments” to “normalize” the data and recreate these charts as well as make some different ones. Any suggestions for different analytics are welcome.