Keep it short and simple for boys
The study, by the Qualifications and Curriculum Authority, found that pupils' performance clearly split along gender lines and adds weight to accusations that last year's test was boy-friendly.
The TES revealed in October that a member of the Rose inquiry set up by the Government to investigate the tests' reliability believed that the rise in boys' reading scores owed more to a "boy-friendly" format than to the literacy hour.
Peter Downes, a respected former head and education consultant, told The TES that the rise in boys' scores in 1999 was largely due to the content and format of the test which involved three short passages about spiders.
He said: "Some questions required a more sophisticated level of response but most of the marks were given for factual comprehension."
He warned that 1999's high results would lead to teachers being blamed for a drop in scores in 2000 if the tests reverted to their usual format.
Boys' reading scores leapt by 14 percentage points between 1998 and 1999. They out-performed girls in a quarter of questions, cutting girls' lead to six points.
The new analysis supports Mr Downes' view of the test, confirming that boys still struggled with open-ended questions which required them to write at length or interpret information.
Boys were good at providing specific inormation in response to clear-cut questions, the report added. They did better than girls in one quarter of questions - but mainly those which "required retrieval of information with a minimum of writing". However, girls demonstrated a better understanding of language features and underlying themes.
The analysis of the 1998 reading test highlighted no areas where boys were superior to girls.
Meanwhile, there is still a huge discrepancy between boys' and girls' ability to spell - particularly among the youngest children. In a 30-word test for seven-year-olds, boys, on average, misspelled six more words than girls. By 11, the gap had narrowed to two words per 100 before widening to 2.6 words for 14-years-olds.
However, the literacy hour, introduced nationally in September 1998, was credited with improving the spelling and punctuation of primary pupils.
In maths, many 11-year-olds were still unable to add and subtract two-digit numbers in their heads, although many primary schools had adopted the numeracy strategy a year before they took the tests.
However, the numeracy strategy did appear to boost children's confidence with mental arithmetic. In 1998, the introduction of the mental arithmetic paper was thought to have deflated overall results by about 1 per cent.
Pupils also had difficulty converting metric unitsand were unsure how to estimate results, use percentages or do calculations involving time.
Children still failed to use their calculators correctly and many tried to tackle questions on the calculator paper in their heads.