The latest combined construction trade survey compiled by the Construction Products Association has thrown up an unexpected and confusing result showing contractors working less in the second quarter of the year.

This is significantly at odds with the official data suggests.

The result of the contractor survey shows 20% more firms did less work than did more in the second quarter. The ONS construction figures point to a rise of more than 6% in output in that period.

This begs the question: surely they can’t both be right?

This question may seem of academic interest and something for the industry statos to quibble over, but, potentially, it could have extremely important implications.

If construction really didn’t grow in the second quarter of this year and in fact shrunk – which is what the latest state of trade survey suggests – this would have a profound impact on the official measure of economic growth – that is the GDP estimate.

It would in fact roughly halve the second quarter figure, which at first stab has been put at 1.1%.

So what? you might ask. Well that would throw plenty of economists and policy makers into a bit of a spin. Though I suspect it wouldn’t alter policy, it would put George Osborne and his team at the Treasury under a bit more strain.

Mervyn King, meanwhile, would probably feel more secure in his judgment that interest rates have to stay low until there is stronger evidence of a recovery and until there is more certainty over the effects of the current Treasury axe swinging.

But the real horror for policy makers would be that such a huge correction in the official GDP figures would chip away at their value as guides to help us through the dark economic forest we seem to have landed ourselves in.

If one thing damages confidence it is uncertainty. So whether they are good or bad the City and the banks of economic commentators want the figures to at least have the suggestion of being grounded in reality.

Let’s not forget the Greek problems were as much related to dodgy figures as they were over its appalling economic mess.

I suspect though – and certainly hope – that such worries are very much unfounded and that there are more mundane reasons for the divergence between the industry state of trade figures and the official measure of construction output.

The first simple explanation might be that this particular survey taken of contractors’ moods is rogue. That happens quite a bit in surveys, especially in times of turbulence.

And it’s true that this survey appears to be out of step with other surveys, including the survey of materials producers within the combined trade survey itself. This shows sales of both heavy and lightside materials growing in the second quarter.

Meanwhile, recent monthly surveys on construction activity conducted for the buyers’ body CIPS also show strong growth over the second quarter.

A second reason for the seemingly out of step results from the contractor survey could be because, being broadly speaking a sentiment survey, it tends to reflect the prevailing mood as much as it reflects the actual performance of companies within the specified quarter.

There may be a tendency for the respondents, to some extent, to look through short-term lumps and bumps and take a more “smoothed” view of their performance.

There will be many other explanations for why this figure appears at odds with others, such as when in the cycle the survey was taken and whether there were any changes to the sample.

But I would suggest, by way of a third, that some of the aberrance might be down to quirks caused by the more severe than normal differential in weather between the first and second quarter.

If contractors were held up through snow in the first quarter and made up for that in the second quarter, it may seem reasonable to them to adjust for this in their responses. The net effect would be to boost slightly the first quarter figure and reduce the second quarter figure.

There are, as we can see, plenty of possible plausible reasons for divergence between the official figures and the results of this survey. However there is no getting away from the fact that the scale and direction of the difference is pretty big.

So how would I read this survey? Well, I’d be cautious about its broad finding that output fell in the second quarter. Too much other data seems to suggest a bounce.

But it does give clues on the relative pain and pleasure within the industry, which is useful. And I would certainly not dismiss it out of hand.

My take would be that, sadly, it adds to a growing pile of evidence, anecdotal and otherwise, that things can only get badder, in the words of that little known pop group N:Ightmare.