Height vs. Length vs. Girth Project

I personally do not believe the average BPEL is as big as the data claims so far, but hey, I just report the data and let everyone else discuss it. I’m not going to alter anything to make it ‘fit’ what I deem correct. I think one problem is due to the fact that people with smaller penises (myself included) are less likely to post. I for one am fearless when posting as the forum is anonymously so who would know it is me? Besides, people couldn’t care less what size you are, we’re all in this gaining game together.

Gimli

I hear you. I’m in the same boat. Heck, I’ve even posted pics of my ‘below average’ unit. Which makes me wonder why the ones with ‘smaller penises’ would be fearful or shy to post their stats on an ‘anonymous’ board?

Ironhammer.

I agree, to each there own though.

Gimli

Gimli,

Isn’t the plot supposed to be from -1 to +1?

Hey man,

I’m sorry but I don’t exactly understand your question. Could you please ask it again in a little more detail? I don’t want to confuse anyone with something I’ve done in the spreadsheet thus far, so if anyone has any questions ask and I promise I will EVENTUALLY answer it. If I don’t in a few days, then give me hell.

Beenthere,
Are you talking about the plots to the right of the spreadsheet, or the correlation coefficients? The plots are just a direct X-Y Variable graph from the database, the line in it is just the linear approximation or “treadline”. Correlation coefficients CAN go from -1 to +1, and for correlation value, one needs only the absolute value ( 1 is ‘perfectly’ matched, -1 is ‘perfectly’ unmatched; man I’m bad at explaining this stuff). Hope one of those was what you were asking, if not, then just post it and I will get to it ASAP. Although I’m afraid I’m going to bed soon, so if don’t get to it in next hour then you gotta wait till tomorrow :) !

Gimli

I thought a proper chart had to indicate -1 as perfectly unmatched, 0 as no relationship, and +1 as perfectly matched. Just reviewing the numbers unplotted indicate correlation, and it matches a -1 to +1 range. I think your range review is flawed. Where’s Para?

You are right with the +/-1 as I mentioned that as well. The plots exhibit very little correlation. I know the linear trendline is sloped upwards, but first understand it is never going to be perfectly level because our dat will not be perfectly normaly approximated. That is why it is best to look at the correlation coefficients while our database is still small and scatterplot is harder to dechiper. The correlation coefficients were created via Excel’s statistical function. I am assuming it is either in the low positives, or the function takes the absolute value. I will check this and see if I can find it in the function definition.

Gimli

Beenthere,

I have looked into it, and I believe that the function is designed to spit out the value between 1/-1. I am led to believe this as I found the equation used for the calculation:

{ Sigma [ ( X-Xbar ) ( Y-Ybar) ] } / {sqrt [ Sigma ( X - Xbar ) ^2 ] [ Sigma ( Y - Ybar ) ^2 ] }

Gimli

I’m from old school and don’t create plots and such like the younger generations. Heck, back when I went to school calculators weren’t even allowed in the classroom. That was considered cheating. And we sure didn’t have Excel. However, I’ve always had a strong knack for math and was always tops in my math and algebra classes including the bit of college algebra I took. That doesn’t necessarily help me here, lol, as I’m just not familiar enough with the spreadsheets. I do know pen and paper are showing correlation, so what’s up?

Unfortunately there isn’t much to go on yet for the pen and paper method. By breaking it down into 66, 67, 68, 69, 70, 71, 72, 73, 74, 75 and 76 inch height categories and then averaging the lengths in each height category I’m showing a progression upward, but there are too few penis sizes in the categories, so the line is jagged in a couple of places. One thing one must think about is that averages don’t do a good job of indicating outliners. The height vs penis poll with about 1300 participants only shows a small upward trend in the average penis size vs height, but the percentages jump up quite a bit when the outliner penis sizes vs height are looked at.

Unless I’m misunderstanding you, then I was correct, no?

To say that a correlation is "significant at the 0.01 level" means that if the null hypothesis were true — if, that is, the two variables were in reality not related — the probability of obtaining a correlation coefficient this extreme by chance would be 1%. (The p-values of ".000" in the table just mean "<.001".)

So far it appears that all the variables here are significantly positively correlated with each other. The strengths of the correlations vary. Squaring the correlation coefficient yields the coefficient of determination , which tells us, basically, the proportion of variability in one of the variables that can be "explained" by the other variable. Take the height-length correlation of (thus far) .451: r-squared = ~0.20; one can "explain" or "account for" about 20% of the variation in penis length by height (or vice versa). This is a fairly modest correlation, but stronger than I would have predicted.

Keep voting, gentlemen!

Here’s a scatterplot of the data (thus far) that provides a graphic illustration of what a correlation of 0.451 means:

Where are all the 4 and 5 inchers at?? :P

The extremes of either scale seem not to be present here.

Yup, it’s incredible how even on an anonymous internet forum guys are shy to report their small sizes or short heights. Come on guys — the average man (in America, where the majority of our forum members reside) is 69” tall, yet only 29% of respondents are this height or shorter?? And only 22% of guys on a PE forum started with a penis 6” or less?!

Remember, you can PM Gimli your data if you don’t want to post them publicly!