A report commissioned by the Office for National Statistics on the cost-effectiveness of early years education, just published, was immediately pounced on by the London-based media, anxious to find conclusions that suited their purposes.
The Times, then Newsnight, seized on a sentence that said "there was no statistically significant difference between total FSP (foundation stage profile, an assessment of five-year-olds' abilities) score for children who attended early years settings and children who did not".
"It's all very complicated," says David Wilkinson, one of the researchers who have seen their document, Quality, Outcomes and Costs in Early Years Education, being brandished as proof that we need to spend lots less - or possibly lots more - on the sector.
He has since made it clear that assessing the value-for-money return of early-years investment is nigh on impossible, given the difficulty of finding out how money is being spent now. More than half of state providers in England, and 30 per cent of those in the voluntary sector, did not know their total outgoings. "It started out with a value-for-money aim, but it just proved impossible within this study," he says.
But, apart from value-for-money, does early years education add value in terms of benefiting children's later education? Mr Wilkinson's study of 7,939 children shows that those who start early education before three do better than those who start later.
Pam Sammons, professor of education at Oxford University and one of the principal investigators in the ongoing Effective Provision of Pre-school, Primary and Secondary Education study, says these findings mirrored hers. "We found that starting pre-school at two years old was better than starting at three," she said.
Spending on early years services in England has risen from pound;1.1bn a year in 1997-98 to pound;5bn a year in 2007-08. There is a debate on whether that is worth it for a rise of 1.5 points out of a possible 113 on the foundation stage profile (now early years foundation stage profile).
But Sian Rees Jones, head of Bognor Regis Nursery School and Children's Centre, says play and exploration are central to how young children develop. She has tracked her children through to reception in various schools and found they do better than their peers.
"It is the quality of the provision that makes the difference," she says. "We have experienced teachers and nursery nurses working in close partnership with parents, a team that has high aspirations and is focused on supporting individual children."
Nonetheless, another report, from the National Institute of Economic and Social Research, reached a more troubling conclusion. In a sample of around 500 children, it found no link between how well children did on the FSP at age five and the quality of the setting attended.
The study also found no link between Ofsted ratings and children's attainments on several different scales, including naming pictures, putting together squares to make a pattern, matching a picture to its pair, social skills and behaviour.
But Colin Coleman, head of Linaker Primary School and Children's Centre in Southport, is in no doubt. He, too, has checked out the pupils who had attended the children's centre at his school since they were aged one or two, and found that they did get higher scores.
He said: "I don't disagree with children being at home with parents, or grandparents or whoever if they get a range of experiences, but a lot of children don't get that".
Mr Coleman added: "The whole thing is very difficult to prove; sometimes you can't prove you are making a difference."