Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
Author
Comment
Giving Sharp Knives to Children

This, I recall, is our bending author's view on providing statistical analysis tools, such as computer programs, to those unskilled in data analysis and statistics.
It therefore attracted by attention to see a white paper on offer described as
"Multivariate Data Analysis for Dummies".
I guess this is intended pretty much like those bodice rippers you take to the beach like "Excel for Dummies", encouraging us all to teach ourselves.... If giving sharp knives to children is dangerous, showing them how to make their own knives might be even more questionable...

But of anyone has read this it would be nice to know.
The paper is downloaded from here:
http://www.europeanpharmaceuticalreview.com/19113/whitepapers/whitepaper-multivariate-data-analysis-for-dummies/?goback=.gmp_1845757.gde_1845757_member_245281671

Given the use to which Doctors and their ilk find for data analysis and its value in the "scare of the month" stories, it is comforting to see that this is being touted by the European Pharmaceutical Review.

Re: Giving Sharp Knives to Children

I don't think J that many people in this forum will be downloading the "Multivariate Data Analysis for Dummies" white paper or book. From what I can see, it looks like a software firm called CAMO Software (based in Norway) is giving away a free book mainly to acquire a load of e-mail addresses and telephone numbers for their sales people to contact. They require a 'valid' e-mail address, which suggests that they may test the validiity of the address before allowing a download. Their main software product is something called "The Unscambler X" and they appear to have a total of 25000 users in 3000 organisations worldwide.

I remember that this multivariate analysis thing did feature in a fairly big junk science story last year. It seems to have been used in the notorious work by Australian psychology professor Stephan Lewandowsky in his paper “NASA faked the moon landing—Therefore (Climate) Science is a Hoax: An Anatomy of the Motivated Rejection of Science”. In a post on his blog Lewandowsky dismissed Steve McIntyre's (of the Climate Audit blog) critique of his paper and retorted that McIntyre "should stop posing as an expert in statistics and take a refresher of Multivariate Analysis 101". It might be noted that Lewandowsky will not necessarily be using CAMO's software as academic researchers often avoid the use of commercial software products.

Incidentally Lewandowsky seems to have been appointed as a professor at Bristol University in the UK. He even managed to get a special grant from the Royal Society, a Wolfson Research Merit Award, which is apparently awarded "to provide universities with additional support to enable them to attract science talent from overseas and retain respected UK scientists of outstanding achievement and potential".