Welcome to the sixth edition of "The Definitive Penis Size Survey"! For those new to this site, the discussion below pertains to the results of the most exhaustive penis size survey conducted to date. Questions concerning everything from actual penis dimensions to the possible correlation between size and other parts of the body, to "owner" satisfaction have been asked in a serious attempt to dissect penis fact from "phallic fallacy". With a combined adolescent/adult sample size in excess of 3,100 subjects, a high degree of statistical precision has been achieved.
In order to expedite the writing process of this communication,
I have followed the same format of previous editions. A number
of topics, however, are covered in slightly greater depth than
before and I have introduced three new graphics.
In a number of the analyses that follow, penis size has been classified
according to the subject's perceived endowment status (ref. question
43 of the questionnaire). The dimensions of this "self-ranked
endowment status" are summarized in the table below.
With the exception of "Penis Size and Age" and "On Circumcision, the analyses discussed below pertain to adult subjects, 18 years of age and older.
The inclusion of minimal, and particularly maximal penile dimensions
have caused many of you to question the validity of my data. Blatantly
fraudulent responses (from 17 year old lawyers or those claiming
"American Zulu warrior" ancestry, for example) are excluded
from entry into the database as are penis lengths in excess of
360mm (14.2"). As an additional safeguard, the bottom 1%
(for those 18 years and older) and top 2% (for all age groups)
of each of the five penile measurements studied have also been
excluded from the analysis. I realize that this is a rather conservative
approach, but the fact that my results are very close to that
reported by Kinsey et al. and the almost symmetric distribution
of erect penis lengths, leads me to suspect that any residual
and unavoidable contamination has been kept to a minimum.
|Please note that all measurements are expressed in metric units with corresponding imperial units in brackets.|
|LENGTH OF FLACCID PENIS||86.3mm(3.4")± 27.0(1.1)||85.2mm(3.4")- 87.4mm(3.4")||157mm(6.2")||25mm(1.0")|
|LENGTH OF ERECT PENIS||161.5mm(6.4")± 31.5(1.2)||160.2mm(6.3")- 162.7mm(6.4")||260mm(10.2")||68(2.7")|
|CIRCUMFERENCE OF ERECT PENIS||127.6mm(5.0")± 25.5(1.0)||126.6mm(5.0")- 128.7mm(5.1")||200mm(7.9")||60mm(2.4")|
|LENGTH OF HEAD||39.5mm(1.6")± 9.9(0.4)||38.5mm(1.5")- 39.5mm(1.6")||76mm(3.0")||12mm(0.5")|
Figure 3 illustrates the bell-shaped distribution of erect penis lengths, thereby
justifying the use of the parametric statistical tests that follow.
Although it was not part of my original analysis, a number of
you have asked whether gay or bisexual men are more endowed than straight men.
As we can see from figure 4, there was no statistical difference in erect penis
length amongst the three groups.
To determine more precisely how you compare with the 2,545 adult subjects who participated in this survey, I have divided the data into percentiles of 10 percentile increments. If, for example, your erect penis measures 155 mm, the table below indicates that this size falls between the 40th and 50th percentile. In other words, at least 40% of the men fell below this value and at least 50% were above. In the upcoming expanded version of this edition, the chart below has been replaced by three graphs which provide an even greater degree of precision.
Since the release of the fifth edition, the number of Black and Hispanic participants has not substantially increased. Because of the continued low sample sizes therefore, the results reported below concerning the aforementioned groups must be considered tentative, pending a larger number of responses.
Surprisingly, it would appear from Figure 5, that contrary to popular myth, Black males have shorter erect lengths than their Caucasian counterparts. However, due to the small sample size and large variation in lengths, this "difference" is not statistically significant. The average flaccid length among Blacks however was 93.8mm (3.7") compared to 87.7mm (3.4") for Caucasians, which may be responsible for the impression that Black men have greater erect dimensions. Again, I emphasize that these observations must remain tentative.
There was a significant difference in length between Caucasians
and both Hispanic and East Asians (those of Chinese, Japanese,
Korean or Vietnamese descent). Whereas with Hispanics, such a
difference was limited to erect length only, with East Asians
all five measurements were statistically lower in comparison to
Caucasians. Although part of this difference may be due to the
shorter stature of East Asians (see "On Correlations"
below), height matched comparisons with Caucasian subjects nevertheless
failed to alter this finding.
The figure to the left summarizes the results of question 44, in which men
were asked to rate members of their own ethnic group with respect to size. Of
particular interest was the response of East Asian subjects. Among East Asians
residing in the East, 31% felt that they were generally of modest size, compared
to 56% of those living in the West. This is probably due to the fact that Western
countries tend to be more racially diverse, allowing for greater interaction
and hence comparison with men of different ethnic backgrounds. It is also interesting
to note that the vast majority of Black subjects stated that they tended to
be well endowed; an opinion which was not borne out by the results of this study.
I was initially surprised by the number of responses from teenagers and, although my survey wasn't specifically designed to detect changes in penis size during puberty, the data collected so far has revealed some rather interesting findings.
As of this edition, and as illustrated in Figure 7, average erect penis length increased until the age of 16, a finding that marks a departure from the five previous editions of this page, which reported that such a developmental milestone occurred at age 17. With respect to adolescent lengths, it is interesting to note that the results presented here are almost identical to that of another online survey, the results of which are located at http://www.dnai.com/~ibex/jackmisc/jacklib/jacksurvey3.html. The apparent variation in average size between the ages of 16 to 49 was due to random sampling error and was not statistically significant..
The situation with regard to erect penis length beyond the age of 50 remains somewhat nebulous. According to the graph below, there was essentially no difference in size with respect to younger age groups. However, I have detected a progressive, yet slight decline in length among older men in this group. Unfortunately, due to the modest sample size of such individuals, it was not possible to determine whether this was a statistical anomaly or an anatomical consequence of the aging process.
The data indicates that during an erection, the average increase in penis size was 197.9%. However, for men whose flaccid penis measured 80 mm (3.1") or less, the increase was 263.2%, compared to 165.3% for those greater than this length (p<0.0001). This principle is demonstrated in Figure 8, which reveals a strong negative correlation between flaccid length and the percent increase during an erection.
It is also interesting to note that there was less variation in erect penis lengths relative to flaccid lengths. This convergence in penis size is illustrated by comparing the coefficient of variation (CV), a unitless measure of variability obtained by dividing the standard deviation by the mean and expressing the result as a percentage. The CV for flaccid lengths was 31.4 whereas in the erect state it decreased to 19.5.
When erect the position of greatest girth was as follows; head 36%, uniform 32%, base 25% and some other location 7%.
Forty eight per cent stated that their penis was straight when erect, 28% reported a deviation to the left or right and 24% claimed to have a bend (presumably in the shaft).
Unlike a previously published study in the medical literature, I did not detect a significant difference in erect length (or the other four measurements) between circumcised and uncircumcised individuals.
Overall, 70% of the subjects were circumcised, 27% were uncircumcised and 3% did not specify their circumcision status. Of the 633 men responding outside of the U.S.A. and Canada who specified their status, 64% were uncircumcised. Not unexpectedly, this compares to 40% of Canadians and only 15% of Americans.
The proportion of those claiming to be content, neutral or discontent concerning their circumcision status was almost equal between the two groups (approximately 66%, 21% and 13% respectively). Among those who were content however, there was a striking difference with regard to sexual orientation. Seventy-two per cent of circumcised straight males expressed contentment with their circumcision status while only 58% of circumcised gay or bisexuals felt the same way (p<0.001).
With the rather impressive number of responses I have amassed since the inception
of this survey, I have detected that, on average, there exists a strong relationship
between erect penis length and height, as illustrated in Figure 9. For the purposes
of demonstrating possible correlations between length and various hand and foot
measurements, it became necessary to control for this phenomena by performing
a partial correlation between length and the various measurements in question.
The results of such an analysis are presented below.
Figure 10 reveals the significant relationship between erect length
The following discussion pertains to Section E of the survey which attempts to assess the degree of accuracy by which penis size is estimated. The results of the first three sets of test subjects are located on a separate page.
As shown in the figure below, subjects A and B were significantly over or underestimated respectively. Although subject A is of modest endowment in terms of length, his circumference is average, and this may have contributed to the overestimation. Subject B's flaccid length was correctly estimated to be 124mm (4.9"), but his predicted erect length, like that of subject A in Set 1, was underestimated. This provides further support that assessing a person's endowment status based solely on the basis of flaccid size is highly unreliable.
As in all other sets, estimates were found to be independent of age, sexual orientation, number of male sexual partners, self ranked endowment status or contentment status.
As in previous editions of this communication, body builders tended to be more endowed than "couch potatoes" (170.5mm or 6.7" vs. 160.7mm or 6.3" p<0.001).
In addition, the difference in flaccid length between those who
are reluctant to potentially expose their penis to public scrutiny
and those who are not remained strong (81.0mm or 3.2" vs.
89.6mm or 3.5" respectively p<0.001). A larger
difference however, existed with respect to erect lengths (166.8mm
or 6.6" vs. 153.3mm or 6.1" respectively p<0.001).
A similar relationship was seen with regard to potential exposure
to male family members.
As illustrated in figure 11, well endowed men had more sexual partners per
year than their average or modestly endowed counterparts. The data reveals that
these men were more self-confident which may, in part, explain this finding.
As illustrated in figure 12, a striking difference existed in the response pattern between gay/bisexual men and straight men when these individuals viewed a well endowed male (ref. survey question 47). Gay and bisexual men tended more to express arousal (36%) or awe (28%), while straight men were more likely to be indifferent (29%) or envious (21%). Among both groups, those who were indifferent had significantly larger erect penises than those who felt either intimidation or inadequacy (p<0.001).
As always, I welcome your comments and suggestions.Richard Edwards