How’s this for Mickey Mouse statistics from Stephen E.Jones BSc, Grad.Dip.Ed?

Yes, sizable error bars are occasionally a fact of life – but there is absolutely no excuse for exaggerating them, simply to score cheap debating points…

Verbatim quote taken from today’s offering  from Stephen E. Jones BSc, Grad.Dip.Ed on his Shroud of Turin Blogspot site.

 “As can be seen, Arizona laboratory’s ages of the Shroud ranged from a low of 561 (591-30) years to a high of 734 (701+33) years, a span of 173 years or 31%. Oxford’s ages were from a low of 685 (730-45) years to a high of 860 (795+65) years, a range of 175 years or 26%. Zurich’s ages ranged from a low of 578 (635-57) years to a high of 794 (733+61) years, a span of 216 years or 37%!

Arizona had the lowest age of 561 (591+/-30) years, while Oxford laboratory had the highest age of 860 (795+65) years, a range of ages of the Shroud between the three laboratories of 299 years, or 53%!

This is a shambles of a result, considering that the three laboratories between them were dating by the same AMS method, the same tiny postage stamp sized sample of the Shroud! And yet they had the effrontery (if not the scientific dishonesty) to claim that, “The results provide conclusive evidence that the linen of the Shroud of Turin is mediaeval … AD 1260-1390″:

“Very small samples from the Shroud of Turin have been dated by accelerator mass spectrometry in laboratories at Arizona, Oxford and Zurich. As controls, three samples whose ages had been determined independently were also dated. The results provide conclusive evidence that the linen of the Shroud of Turin is mediaeval … The age of the shroud is obtained as AD 1260-1390, with at least 95% confidence” (Damon, P.E., et al., “Radiocarbon Dating of the Shroud of Turin,” Nature, Vol. 337, 16 February, 1989, pp.611-615, p.612).

Is that how they taught you to handle scatter in experimental data when you did that BSc of yours, Mr. Jones?  First take the lower boundary of one of the lowest of the triplicate readings, then take the higher boundary of the highest of the triplicate readings. Subtract the lowest from the highest to get the highest possible value for the range, and then, wait for it, calculate what that range is as a percentage of the lowest value to get the least flattering result possible for experimental error.

This is not statistics, Mr.Jones, or if it is, then  it is Mickey Mouse statistics. This is to treat data with the same contempt with which you treat people (one in particular). What a singularly mean-spirited person you are Mr.Jones, and tad hypocritical too, some might think – in bandying around that term “dishonesty”.  Ever heard of the term “intellectual honesty”, Mr. Jones – something you clearly lack in the cavalier and shoddy way you cobble together this and your other partisan arguments, all the while flaunting those so-called academic credentials of yours.

Here’s a more reasonable statistically-informed way of looking at the data.

Arizona:    range:  561-734  or mid-value of 647.5  +/- 86.5   

Max deviation from mid-range value as a percentage of mid value is 13.4%

Oxford:  range:  685 -860  or  mid-value of  772.5  +/- 87.5

Max deviation from mid- range value as a percentage of mid -value is  11.3%

Zurich:    Range: 578 – 794  mid-range value: 686 +/- 108

Max deviation from mid-range value as a percentage of mid value is 15.7%

Even without the standard deviation (SD) , i.e. using (as here) non-standard deviations due to lack of replicated data, it is conventional – to say nothing of reasonable – to calculate deviations from a mean as a percentage of that mean.  Serious scholars with a reputation to maintain do not go lampooning and devaluing others’  data in the crude propagandist way that you have done in your travesty of statistical methodology, Mr. Jones. But then you are not a bona fide scholar – are you Mr.Jones? You just masquerade as one while single-mindedly pursuing that bible-bashing agenda of yours.

You say you are a supply teacher. Methinks it is YOU who needs to be back at school, facing the teacher and the white board, and doing your homework properly, especially Statistics 101.

 Finally, to restore some fairmindedness, to say nothing of sanity, let’s imagine that the Shroud really were early 1st century AD, say 2000 years old, and that it had been radiocarbon- dated with the same error values we see above.  (This has to be an approximate exercise, needless to say, in which for simplicity and convenience I have taken a nice round 2000 years as the “real”age).

Arizona  would have reported 2000 +/- 13.4%  ie.  (2000 – 268)  to (2000 +268)  i.e. 1732 to 2268 years old

Oxford would have reported 2000 +/- 11.3%  i.e  (2000-226) to (2000 +226)  i.e. 1774 to 2226 years old.

Zurich  would have reported 2000+/- 15.7%,  i.e. (2000-314) to (2000+314)  i.e. 1686 to 2314 years old.

(Again, this is simplified, since I have centred all three on the same 2000 year figure. In practice, the individual values would show a distribution around that value,  say +/-200 years, but with virtually zero probability of an overlap with the medieval Shroud estimates, given the staggering difference in age – see below).

Now let’s calculate some grand means for all 3 laboratories for the actual age of the Shroud, and compare that with the age that Mr.Jones and his fellow authenticists had wished it to be:

 Actual Shroud:  702 years old   (+/- 13.5%)

 Fantasy Shroud: 2000 years (+/- 13.5%)

A conventional representation of those error bars (schematic)  – a far cry needless to say from Stephen E. Jones’s Mickey Mouse statistics

So your fantasy Shroud  works out as (2000-702) years , ie, 1298 years older than the real one, which is 2.85 times as great,  or 185%  greater in relative terms.  These are the figures that matter, Mr.Jones. not your Mickey Mouse statistics.

Postcript: this post addresses the data as b*ggered by Stephen E. Jones. It does not attempt to address the issue of the size or representativeness of the sample(s) taken for analysis, which is a separate issue  – one that I may return to another day. Suffice it to say that I regard all the talk about “repair by invisible reweaving”  with “end-to-end splicing and dyeing ” of individual threads as desperate special pleading, or as I prefer to call it: Mickey Mouse science.


About Colin Berry

Retired science bod, previous research interests: phototherapy of neonatal jaundice, membrane influences on microsomal UDP-glucuronyltransferase, defective bilirubin and xenobiotic conjugation and hepatic excretion, dietary fibre and resistant starch.
This entry was posted in Shroud of Turin and tagged , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s