Statistics tell us nothing without context.  And each person's statistics tells the reader exactly what the author wants to them to hear - i.e. whatever the reality is, statistics presented in a particular way can say whatever you like.

We have had a few complaints about our surveys.  But we always put our results in context.

Whereas the people that want to knock Microsoft (and particularly Vista) come up with articles like this one below which was posted on Silicon.com and is based on a survey by the Corporate IT Forum (Tif).   It says things like 35% of people put themselves in the "not yet interested in Vista" which is about what most surveys suggest.  So fine.  But then they put a figure of 30% of respondents as "currently investigating Windows 7" compared with only 14% for Vista.  Well that might be because many people have already "investigated" Vista and decided on their strategy and not because, as implied, most people intend to bypass Vista.

My other gripe is that the article doesn't say how many people responded to the survey.  If it's 1000 across a range of organisations it means one thing.  If it's 250 UK SMEs it means another.  In fact it was 43 Corporates from across the world.  Which for me means that it is worth diddly squat.  Looking at the Tif website their view of their membership is that "many are Fortune 500, Euro 500, FTSE 100 or FTSE 250 listed" none of the figures are statistically significant across that range.  I know it's only Tif members that are asked to respond but would I base my corporate strategy on the response of 43 organisations from around the world ?

Oh and the other big mistake is to mix statistics from different surveys.  So the bit about Vista uptake among businesses was slow during 2007, although a quarter of businesses said they planned to upgrade in 2008 is slightly more significant as that survey by Forrester (a slightly better known name in the research world) quizzed more than 50,000 users from 2,300 large enterprises over the course of 2007.  So 25% of 2,300 enterprises is 575 intending to move to Vista in 2008.  And Forrester's research found that 30% of respondents had IE7 whereas the Tif survey has only 4% of respondents.  That's a huge difference, and the Forrester survey was covering 2007 so one can reasonably assume that the percentage figure has increased.

So no conclusions can be drawn from the survey and news item, other than the author and editor think that knocking Microsoft is news worthy.

Microsoft's Windows Vista operating system is still playing second fiddle to XP with business users, with more enterprises confessing to checking out the unreleased Windows 7 OS than its predecessor.

More than half (58 per cent) of businesses using Microsoft technology are "exploiting" Windows XP compared to just four per cent for Vista, according to the 'reality checkers' research by the Corporate IT Forum (Tif), seen exclusively by silicon.com.

Tif's corporate IT reality checkers survey helps its members quickly compare the progress and position of their companies' IT against the technology choices of other members.

Tif also found that 35 per cent of organisations describe themselves as "not yet interested" in Vista.

The OS most people appear to be developing or piloting is XP, with 12 per cent of businesses saying they were doing so compared to five per cent for Vista.

Interestingly more businesses said they're currently investigating or analysing Microsoft's next scheduled OS, Windows 7 (30 per cent), than Vista (14 per cent).

In contrast, seven per cent of businesses even said they're still exploiting Windows 2000 although 19 per cent said they are currently replacing or "sunsetting" it.

Back in April, research showed Vista uptake among businesses was slow during 2007, although a quarter of businesses said they planned to upgrade in 2008.

The main reason given by Tif members for not moving to Vista was a lack of business requirement to do so.

Meanwhile, Microsoft's latest browser, Internet Explorer 7 is having a similar battle with its predecessor IE6, with a fifth of respondents saying they're not yet interested in the newer version of the app.

Almost two-thirds of businesses surveyed (65 per cent) said they are exploiting IE6 compared to four per cent for IE7. However, 14 per cent said they are currently piloting IE7 with the same proportion using it in isolation.

Almost a quarter (23 per cent) said they are analysing and investigating IE8 which is currently available in beta form.

Again on Silicon.com (this isn't a vendetta against Silicon.com, it's just that I regularly read it) they are conducting a survey on Browser usage. The problem with the results is that, again, you don't know how many people responded.  If it's 100 you would view the result differently than if it were 10,000.

Internet Explorer     27%

Firefox   53%

Safari   5%

Chrome   8%

Opera   6%

Flock   0%

Netscape Navigator   0%

When we produce our surveys and quote percentages you are aware of how many organisations are in the survey, how many pages of the websites surveyed we looked at and you know which organisations are in the survey. That way you can make judgements on the figures we produce and quote.