Sunday, October 6, 2013

Erect penile length and circumference dimensions: a new internet survey - Need for a wider range of condom sizes

A new research study from the Indiana University research team was published online some time ago in the Journal of Sexual Medicine

Like many studies of penis size, they relied on men to report their penis size.
Then should we be concerned that men would lie about their size? No. Not really. 
Dr herbenick answers: 
"In fact, unlike most previous studies of self-reported penis size, they had good reason to report accurate data to us because we were using their size data to match them to a condom that was sized to fit their erect penis. If they reported a bigger-than-reality size to us, they would get a baggier condom. If they reported a smaller-than-reality size to us, the condom would be too tight."

  •  Measurement
It was performed using two erect penile measurement tools (one that used a letter-coding measurement system and a second that consisted of a centimeter-based measurement system) and detailed, illustrated directions about how to measure their erect penis, from the underside base and choosing the letter or numerical code that is “closest to the end of the head of your penis,” for the purposes of the study.

You will notice that the measurement is taken from the underside of the penis and therefore, can not be compared to the ususal measurement procedures.

  • Results
Below are the results obtained from this study and largely communicated to the media:
NB: values are in cm. Divide the value by 2.54 to get the equivalent in inches.

  • Discussion
I would like to spend some time about the concept of mean and standard deviation.
According to these raw results, one should expect that:
  1. 50% of men have an erect penis girth of less than 12.23cm.
  2. 84.13% of men have an erect penis girth of less than 14.46cm (mean + one standard deviation).
Provided that the data are normally distributed.
In fact this is absolutely not the case. Therefore, the standard deviation (2.23 cm) does not provide an accurate percentile rank. The real data give the following repartition:
  1. 50% of men have an erect penis girth of less than 12cm (this is obvious - the median was given).
  2. 84.13% of men have an erect penis girth of less than 13.5cm.
  3.  14.46cm would be at about the 93rd percentile!
The chart below give the detailed distribution:

  • Reminder
Normally distributed data looks like a bell curve. One has  to assume that penile dimensions data are not normally distributed because penile dimensions can not be smaller than a minimal value (in order to be physiologically functional), but the upper limit can be very high (before impacting negatively the functionality). 
The direct consequence is that the theoretical bell curve symmetry is broken with more individuals below the mean and grouped in a smaller range while fewer individuals are above the mean and distributed in a wider range.
When graphed, this type of data looks like a bell that leans to the left. 
In a effort to show it, I have gathered few "self-reported" studies, known to be "as serious as possible" scientifically. The collected data may not be as accurate as the one performed in a laboratory, but the phenomenon is more obvious as the number of tested individuals is important and therefore better reflects the broad range of penis sizes.
The selected studies are:
I have created a graph which compares these four studies to the few meaningful percentiles used when the curves are normally distributed:
  • the 50th percentile (the mean when it is normally distributed, the median otherwise)
  •  the 15.87th and 84.13rd percentile respectively representing the mean (M) - one standard deviation (SD) and the mean + one standard deviation when the distribution is normal
  • the 2.28th and 97.72nd percentile representing the M-2SD and M+2SD when the distribution is normal
  • the 99.87th percentile representing the M+3SD (the M-3SD was too small to mean anything) when the distribution is normal.
Graph: Cumulative percentile data of four erect penis circumference studies in cm compared to the usual normal standard deviations percentiles.

  • Conclusion
These data help showing once again the importance of having a wide choice of condom sizes, in order to better reflect the wide distribution of penis sizes.

According to the Herbenick study, as many as 15% of the male population has an erect penile girth of more than 13.5cm (5.3in). For this group, a normal sized condom is stretched 30%, which is really uncomfortable (for some, even painful) and accounts greatly for the high fail rate of condom usage.
The same can be said of the 15% of the male population whose erect penis girth is less than 10cm (4in). This group simply can not wear a normal condom without experiencing severe slippage.
Condom producers have a great responsibility in proposing the greatest possible range of condom sizes. Producers like Theyfit, My.Size or Coripa are more than welcome.

Sadly, sometimes, the governments are the cause of the lack of a wide choice. The best example is the US themselves which forbid the condoms to have an opening of more than 54mm (un-stretched circumference of 108mm). Let's hope this will change in the near future, thanks to people like Dr. Herbenick.

Debby Herbenick, Michael Reece, Vanessa Schick, Stephanie A. Sanders. Erect Penile Length and Circumference Dimensions of 1,661 Sexually Active Men in the United States.
The Journal of Sexual Medicine.

Other references:
Our New Research on the Penis Sizes of 1,661 American Men
Psychology today 
NB: To receive a copy of the full article from this penis size study, email Dr. Herbenick at cshp (at) indiana (dot) edu.
In the article you will be able to see some very interesting charts like:
  • Participant socio-demographic characteristics by penile length and circumference
  • Measurement and erection characteristics by penile length and circumference
  • Some discussion on the possible future research. 


  1. unfortunately the method is not 100% lie proof...since there will aways be those who don't go to the study willingly but are only dragged by friends or challenged/dared by colleagues these would be concerned about reporting the result to their friends and thus lie etc....
    and those who never use a condom or already find the normal condoms perfect for them , this category wouldn't be so enthusiastic about receiving a condom and thus prone to mess with the reported data....obviously they would give bigger numbers....

    so for me the smaller numbers are more solid than the bigger numbers .....which implies that the medians are even less than what is found in this study.....

  2. Dear blogger, thank you very much for such a thorough blog about this topic about condome sizes. I was just searching on a jobrelated issue about condome, will gladly return later.

    Alexander from

  3. I was looking for something like this ..Really an nice and informative post..

    Tips for Levitra

  4. I have a really hard time believing these averages. I consider myself pretty skinny at 13 cm of girth.

  5. +aren't you making mistakes in your analysis? If 15% has a girth of 14cm + 5 % 15 cm that means at least 20% has a larger size than 13.5, since the numbers were rounded up?

    1. Think twice:
      Roughly 74% are 13cm or less.
      Rouhly 90% are 14cm or less.
      With a straight line between these two points, you would have 74% + (90%-74%)/2 = 82%.

      But it is not even a straight line, as there are more individuals between 13 and 13.5, than between 13.5 and 14cm (nature of distribution).

      You got it?

    2. I'm afraid I still can't comprehend because "participants were rounded up to the nearest number", that means the 23% that have 13cm of girth range from 12.5 to 13.4? Same thing with the other numbers.

    3. So it should be: roughly 74% are 13.4 or less, am I right?

    4. Yes you are right... But doing so would contradict the 12cm median (it would become 12.35cm). So I have chosen to stick to the midpoint methodology which will cancel itself out over a large sample (it may not be the case here).

      Side note: the fact that the ruling system is made of numbers that do not correspond to the centimeters. The midpoint methodology is used, for every number and then rounded to the closest centimeter.
      For example, the number 17 corresponds to a girth from 135 to 141mm. It is OK because every measurement is allocated to the midpoint 138/139mm and this is then rounded to 140mm. But consider number 21, corresponding to 142 to 152mm. Every measurement would be allocated to the midpoint 147mm, rounded to 15cm. This is not right, because a part should have been allocated to 14cm (from 142 to 144mm).
      Same goes with number 99 (123 to 128mm).

    5. So in your honest opinion, 13cm is nothing to worry about? There are more studies that have come up with a median around 12, but they used an erectiometer which gave a more conservative reading of the actual size.

    6. Read all the studies, the serious ones, and you will accept the facyts.
      More important is that your significant other will love you for what you are, not the package you carry.

  6. This comment has been removed by a blog administrator.

  7. Are these measurements taken mid-shaft? Because some studies you've quoted like the one from Harding (2002) was taken just below the glans (the thinnest part of the penis). Don't know about the other studies, but they always vary a lot.

  8. Thanks for sharing useful information about maintain sexual health. vardenafil online is best for men's health.

  9. I saw a chart similar to the one above for length, but now can't relocate it. Help?