This is an article about a statistic

(Inspiration here.)

A numerical variable was today reported to have either increased or decreased or remained constant. Depending on which one, this may represent a record value for this variable, a dramatic rise or fall since whichever point in time is required to show a dramatic rise or fall, or nothing whatsoever. In a development which is probably entirely unrelated, although there is no way the business editor will admit this and publish this bit, the FTSE-100 share index rose or fell slightly on the news.

Speaking to this website, a spokesman for a lobby which claims to care about the current value of this variable said that the integer demonstrated clearly that the lobby was right. The spokesman said that the government must act, that the government must immediately stop acting, or perhaps that the current value of the variable showed that although insufficient, the government’s policy was a step in the right direction. In any case, it demonstrated the enduring relevance of their members’ concerns.

Reached for comment, the Ministry of Variable said that it was going to take tough action on the number. Friends of the minister said he or she fully understood their concerns, but that he or she would not be stampeded into action. However, the minister will say, modernisers would not be held back in the comfort zone by variable interests. The Shadow Minister for Variable said that the government was relying on a fundamentally flawed measurement and that their own preferred measure showed that the variable should be significantly higher or lower. He or she accused National Statistics of twisting their measurement of the variable to suit the government of the day. The Minister’s office retorted that they would take no lessons on variable from a party that had allowed the value of variable to rise to record levels, fall to record levels, or stagnate at a constant level when they were last in office.

The Campaign For or Against Variable said that the public were in danger and the precautionary principle should be applied. “So-called statisticians claim that this level of variable is perfectly safe, but how can anyone really know? Also, the weekly level of variable has been recorded as being as high or low as X in the last six months, when the statisticians say there is only a 5% chance of this. So how come it’s happened once in 24 weeks?”

The mean value of variable over the last 20 years is Y, and it typically varies as much as Z year-on-year. Over the long term, variable in the UK is typically A% higher or lower than the average of OECD countries, EU countries, or the world. On the basis of variable’s distribution and standard deviation, this week’s value could be expected about every B years, and therefore this news is either important or pure noise. To be more accurate, the variable should perhaps be given as a percentage of GDP, as a per-capita value, a median value rather than a mean, or as a percentage of some total or broader average. It’s very likely that this may be explained better by drawing a graph. Unfortunately, this paragraph was edited out of the final article, or quite possibly the author never bothered to write it in the first place.

Both the people we spoke to in the street, because they looked likely to say something sensational, who recognised variable but didn’t bother us with comments like the paragraph above said they were deeply frightened and baffled by the issue.

  1. Neil

    30 seconds after this report is published, Tim Worstall posts a blog entitled “Errr, you were saying, Polly?”, containing a blockquoted paragraph pasted from the report and the words “Well yes, quite.”

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 )

Connecting to %s

%d bloggers like this: