----------------

Not Very Dear Editor :

1. The ONLY reason we continue to subscribe to the Herald is that there is no practicable alternative. Around 80% of it typically goes straight into the recycling cupboard; another 15% goes in meaningless pictures. ( Yes, the numbers are guesswork; I can't support them with real measurements. I have intended to make such measurements for years, but I have always found better things to do. In the light of what follows, I point out that when I quote statistics I include an assessment of their reliability. ) We read the other 5% in the hope of finding some grains of news among the detritus of illiterate and innumerate drivel.

2. And we find some. That's just as well, because the Herald is our only source of news. We gave up television many years ago because it was even worse and wasted far more time than we had to spare, largely because we couldn't skip over the drivel. Radio news is perhaps better, but takes longer, and comes at inconvenient times. I record that, despite my annoyance, the Herald has kept us pretty well informed. The material is there.

3. That said, it seems to me that your quality is getting worse, and not slowly. The standard of English is deteriorating; many of your writers seem unable to manage elementary spelling and punctuation, and no one else notices. Worse, there is little evidence that anyone on your staff has any understanding at all of mathematical, scientific, or technological matters.

4. - all of which is a prelude to my immediate concern, which is an article in today's "Weekend Herald". On page A18 you have printed a thing under the heading "The long and short of heart disease", accompanied by a totally irrelevant picture. It is an example of the "medical" "news" with which you frequently pad out your pages. This is usually material which I skip, but my wife is short in stature, so for once I thought there might be something interesting.

5. The first part is plausible, if rather boring. In the last six paragraphs and the green panel, this changes dramatically. We learn that :

• - A survey covered 4000 women.
• - 700 had heart disease; ALL of these women's legs were between 74.6 cm and 75.3 cm long.
• - ALL the legs of the other 3300 women were between 75.8 cm and 76.0 cm.

6. Are we really to infer that :

• - Practically ALL British women's legs are between 74.6 cm and 76.0 cm long ? There is no mention of any selection, so we must assume that the sample is random. ( NOTE : Perhaps that's why we settled in New Zealand. It is not clear precisely how the legs were measured, but my wife's legs appear to be less than 74.6 cm long, so she is clearly not an acceptable British woman. )
• - The legs of over 80% of British women are between 75.8 cm and 76.0 cm in length - a range of just 2 mm ?
• - Despite the vast number in that range, there are NO women with legs between 75.3 cm and 75.8 cm in length ? That surely suggests two species.
• - The leg length is an absolute determinant of a British woman's destiny so far as heart disease is concerned ? Those with legs shorter than 75.3 cm in length are doomed !
• - Lung function, smoking, and social status have absolutely no bearing on the issue ? ( Yes, earlier in the article a correlation is suggested at the level of "more likely", but the correlation with leg length is presented as perfect. )

7. No, we're not "really to infer that", though that's what your article implies. In the last paragraph, we learn that :

- The risk falls by an average of 16% for every extra 4.3 cm in leg length.

That's almost believable as a real, if oversimplified, result. Even so, the mention of 4.3 cm remains puzzling - as the overall range of women's leg lengths studied is 1.4 cm, how can they tell ?

----------------

Given such ridiculous material, it's easy to poke fun. The real question is why it should ever have been printed. Isn't it obviously so remarkable that it must be either nonsense or a new discovery that calls into question much of our current understanding of physiology ? Does no one on your staff have any responsibility for assessing the plausibility of your material ?

I have wondered before, but this rather extreme example raises the question again. Some of the material you have printed about ( for example ) President Bush seems almost equally bizarre when assessed as probable behaviour of a world leader; can we discount that too ? Please ?

           Alan Creak.

----------------------------

P.S. : In the unlikely event that you want to print this, you're welcome, but I would much rather someone took some notice of it and did something about the problem.

- and for the record :

Alan Creak,
2004 June.