Price increases vary depending on where you are, what's nearby, what street your on and even what sort of house you live in.
Yes, many valuations are done by 'comparables' (i.e. seeing what else sold nearby) but as EA's we have to take into account other factors;
Is it South facing? North facing? Is this house the same size as the others on the road that we're using as comparables? How long ago was the last recorded sale, and how much have values changed since that date. If the data is old, we need to go further afield for a comparable. Is the décor good? Are there any structural problems? Is it attractive to the type of buyer looking in this area? What does this house have that others do not, or what does it not have that others do?
There's a great deal of 'compare and contrast' with valuing that has an influence on prices, but it's safe to say that demand massively outstrips supply and that will keep prices buoyant on it's own. I work in an extremely high capital growth area of Hertfordshire and that is due to it's extreme desirability. Other areas will fare differently, but then the factors which impact the value will also be different.
The reason for all that waffle is that it's easy for national stats to say "house prices have increase b x%" but that could mean that they've gone up by 8% in Richmond and have fallen by 2% in Stoke on Trent - it's a bit of a non-event as house price trends tend to be hyper-local.
Take a look at home.co.uk and punch in your postcode to see house price performance stats. It's very interesting. Also try Mouseprice.com and the land registry website. That's where the hard data comes from.