Some primaries are spending up to nine times as much as other schools on literacy work with infants. The per-pupil staffing costs can range from pound;120 to pound;1,025 a year.
The amount of time devoted to reading and writing at key stage 1 also varies widely despite the introduction of the literacy hour.
Some schools dedicate less than four hours a week to
literacy teaching while others set aside up to 15 hours. The average is 5.66 hours.
These surprising variations have been uncovered by a survey of 534 schools in 125 English education authorities carried out in spring this year.
The study, funded by the Open University, suggests that the Government's National Literacy Strategy has left schools with more autonomy than anyone anticipated.
The spending discrepancies stem largely from class-size
differences - a tenth of Reception and Year 1 classes and 17 per cent of Year 2 classes had more than 30 pupils. But costs are also affected by the number of hours that a school chooses to allocate to literacy work and the salary bill for learning support assistants.
Three per cent of schools set aside less than five hours a week for literacy, while 5 per cent devoted eight hours or more (see table).
On average, each infant class received four hours' literacy help ach week from learning support assistants. But there was no consistent pattern. Nine per cent of schools said assistants spent less than two hours a week on
literacy while 13 per cent
enjoyed eight or more hours'
The average staffing cost was pound;295 per pupil - 15 per cent of schools' revenue. Most of this sum (pound;204) was for teaching the literacy hour. The additional costs were other teaching time (pound;29), general learning support assistants (pound;40), literacy co-ordinator (pound;11), and special needs assistants (pound;11).
"Schools tended to spend more on literacy the higher their revenue per pupil," the researchers say. "But there was still a considerable range of spending for any given total revenue per pupil."
The amount of help offered by learning support assistants also seemed unrelated to schools' revenue, class size or catchment area.
The researchers acknowledge that schools serving poorer neighbourhoods tended to have smaller classes, but add that extra cash is not being systematically directed at children who are "struggling to become literate".
Most heads were satisfied with the quality of learning resources. But they said they would ideally like to spend more on small-group teaching, learning support assistants and the least able children.
"The Resourcing of Literacy at Key Stage 1", by Rosalind Levacic, Institute of Education, London University, and Derek Glover and Megan Crawford, Open University. The report is on the web at www2.open.ac.ukheadswhatsnewrolps.htm