This week’s high profile spreadsheet story was the FT’s savaging of Thomas Piketty. These errors have raised questions around Piketty’s (until now) widely praised conclusions, regarding the rising wealth inequality in the US and Europe.
The story has received commentary from a few sources, including:
Felienne Hermans’ blog – Felienne Hermans is an assistant professor at Delft University of Technology and her attitude is “only a fool blames their tool”. Whilst spreadsheets are very close to her heart – she has a wider view about programming and software engineering methods.
In my opinion the spreadsheet blunders that are mentioned in the news are as a result of a lack of respect for the tool of choice (the spreadsheet).
To look at this a slightly differently, it is like an average person like me, who has no building skills, looking at a house and thinking “I could build that”, simply because I have the right materials.
If you don’t have an understanding (or even awareness) of the basic principles and standards for building a house, then it will not work. You might not realise there is a mistake, but that house will always be fundamentally flawed and dangerous.
But as so many spreadsheet errors have shown, a poorly built model doesn’t always fall apart in the same way a poorly built house may. The mistakes can remain hidden until someone with an understanding of the standards takes a look, then the mistakes appear glaringly obvious.
Themodelauditor blog points readers towards the ICAEW’s twenty principles for good spreadsheet practice. This is due to be launched formally launched on the 17th June in London.
Principle number 2 in this document, is “Adopt a standard for your organisation and stick to it”. We’ve been talking about this for some time and we’re delighted that a body such as the ICAEW is clarifying its advice on this.
At F1F9 we use the FAST Standard. This helps us to ensure that any errors that do exist (and they will) are more easily spotted.
Until standards are embraced within spreadsheet use, mistakes will continue to occur and the hard work that has been put into developing hypotheses and conclusions will be undermined by simple errors that could have been avoided.
Almost a year ago we highlighted what happens when spreadsheets go wrong, with our dirty dozen ebook – take a look and tell us if you think that Piketty should be added to make this a baker’s dozen.
