The latest quarterly English Business Survey produced by the business department BIS adds further weight to the notion that the construction market is improving.

The survey uses a weighed balance and showed 30% more construction firms saying workload picked up than saying workload shrank between the first and second quarters of this year. Given that about a third saw workload stand still that is an impressive majority.

Other highlights include a 26% positive balance expecting to see workloads increase next quarter, a 21% majority saying more work than a year ago, a 15% majority saying more employment than in the previous quarter.

Meanwhile a 12% positive balance said labour costs went up and a 13% positive balance of firms expect them to rise again in the coming quarter.

On the face of it this all sounds quite categorical. It suggests we had a pretty strong improvement in the industry in the second quarter.

But what does it all mean in practice? Let’s look at what the hard statistics can provide us with as a guide to what actually happened.

For workload we have a problem because we are comparing GB statistics with English Statistics. But what we find is that construction output on a volume seasonally adjusted basis actually rose in the GB by just 1.9% in the second quarter.

That’s not a great deal, but noticeable, although how noticeable is a moot point given that the industry has been bouncing up and down by that amount quite a bit of late.

If, however, we look at the current price non-seasonally adjusted data suggest a rise of 8.2%. That’s quite a lot. And a 7.1% rise if we take account of inflation. So if we are looking at cash things were up a fair bit.

But if we look at the volume data comparing this year’s second quarter with a year ago we see a rise of just 0.4% and 0.3% if you use the non-seasonally adjusted figures. That is hardly noticeable. Ignore inflation and we see a rise of 3.1%. That will be noticeable, but so will the effects of inflation.

Now let’s look at employment data. Here you can get workforce jobs numbers just for England. These show a rise in the total English construction workforce of 0.5%. Not a lot. But what we also see is that the number of directly employed is rising slightly faster than the self-employed numbers.

Looking at the output and the employment figures in the round my view would be that we are seeing an industry pretty much flatlining with a hint of improvement.

That seems a bit at odds with how we might interpret the English Business Survey results.

Well yes. But this is to be expected. Here’s just one reason why we have to be cautious with our interpretations of what are really sentiment surveys. I’m not sure all those filling in the questions pore over their books to find exact answers, my gut feel is they use their gut feel. So if they feel better they will shade up, if they feel worse they will shade down.

That doesn’t make these surveys false or bad. In fact in some ways you may be getting a different take on what’s happening in the real world through the change in sentiment that adds texture and helps your understanding.

There are of course other things that influence these surveys. Do the respondents look at the cash or the amount of work? Is there optimism bias? Is there survivorship bias effects?

Probably all of those things come into play, aside from the statistical and methodological issues associated with the survey itself.

But the point to note is that construction folk feel a bit happier about their lot. That’s a good thing.

The trick, however, is not to get carried away with sentiment surveys and end up disappointed when things are not as rosy as they are portrayed.