A-level results day can sometimes feel like Groundhog Day: the same story greets you over the breakfast table year after year. Every August, we learn that A-level passes have risen to another record high, sparking an "intense battle" for university places.

According to the headlines, up to 250,000 applicants could miss out on university this year, while thousands of state-school pupils are being entered for "soft" options at A level to boost results. But is the picture really so gloomy? What do the data show? Here we consider whether the claims and assumptions made about A levels stand up to scrutiny.

'You need A levels to do a degree'

The media's obsession with A levels carries the implicit assumption that they are the be-all and end-all when it comes to university entry. Not so: in fact, almost half (48.5 per cent) of applicants accepted on to courses at UK ÖйúAƬ institutions via the Universities and Colleges Admissions Service in 2009 did not have A levels at all (see related file, right).

Alongside the significant proportion of applicants holding overseas and Scottish qualifications, students are increasingly taking advantage of a range of vocational qualifications that can allow entry to degree courses. National Diplomas, Key Skills and OCR Nationals are just a few of the qualifications attracting Ucas tariff points, and 11.4 per cent of successful applicants were accepted with only BTECs - more than double the rate 10 years ago.

ÖйúAƬ

Kate Westmacott, qualifications information review project manager at Ucas, says: "There is more variety out there and students have increasingly moved to a range of qualifications. For instance, 67.6 per cent of students applied with just A levels in 1999, but that had fallen to 49.8 per cent by 2009."

The staggering array of Level 3 qualifications (equivalent, in theory, to A levels) now presents a headache for Ucas, which is reviewing the weighting of courses.

ÖйúAƬ

'More students are taking soft subjects'

The decline of "proper" A levels is much lamented in the media's post-results coverage. Alongside the obligatory photographs of confident 18-year-olds celebrating their straight-A sweeps, there are inevitably a couple of paragraphs bemoaning the march of media studies, communication studies and PE.

Well, yes, it is true that media studies is still going strong. A total of 33,375 sixth-formers took the subject last year - but that was down 1.3 per cent on 2009. (Interestingly, only 12.5 per cent of candidates achieved an A* or A in this so-called soft subject, compared with the 44.8 per cent of pupils who gained the same grades in mathematics.)

But it is the traditional "tough" A-level subjects that have proved most popular in recent years. For example, the number of further maths candidates jumped 11.5 per cent last year (up to 11,682 candidates) - the highest percentage increase for any subject in 2010. This was closely followed by economics and maths, which enjoyed rises of 9 per cent and 6.2 per cent respectively.

The sciences also enjoyed renewed interest: the number of A-level physics candidates rose 5.2 per cent to 30,976; chemistry was up 3.7 per cent to 44,501; and biology had 4.3 per cent more students at 57,854.

Libby Steele, head of education at the Royal Society, says recent rises have not yet made up the long-term drop in science and maths candidates. "The numbers are starting to improve and we welcome this. But you have to look beyond them to the percentage of each cohort taking maths and science. More students are taking A levels, but the numbers studying maths and science have not increased proportionately."

Steele explains: "For instance, 12.8 per cent (84,744) of students took maths and further maths in 1989, but that dropped to 7.5 per cent (55,917) in 2003. Last year the figure was 10.4 per cent (88,683), which is a lower percentage than 1989, although the real numbers have gone up slightly.

"The sciences have also had their peaks and troughs. In the late 1980s and early 1990s, we had quite high numbers doing chemistry, biology and physics. The numbers then dropped significantly for chemistry and physics until they started to go back up in 2005-06. But physics has still not recovered to where it had been in real terms or as a percentage of the cohort."

But she adds: "We are pleased that more people are taking maths with science. This means that the message is getting out to students that maths is an important subject, both in its own right and as a prerequisite for science degrees."

ÖйúAƬ

Compared with the mini-revival in science, technology, engineering and mathematics - or STEM - subjects, students are actually shunning those courses derided by some parts of the media as easy ways to get grade As.

The biggest loser last year was critical thinking, the number of candidates for which dropped 16.4 per cent to a mere 2,082, while the numbers doing PE and communication studies also fell.

Even general studies - seen by some critics as an easy way for sixth-formers and schools to bump up their Ucas-point tallies - is declining in popularity. From a peak of some 60,000 candidates in 2006, demand has plummeted, with 46,770 students sitting the examination last year.

'It's harder than ever to get a place at university'

The apparently ever-improving A-level results are no guarantee of access to ÖйúAƬ. More than 210,000 would-be students were left without university places in 2010, the headlines screamed last year.

With limited places and growing demand, who could not feel for the legions of young people missing out on tertiary education?

To illustrate the shortage of places and the "unfairness" of the system, stories were often conflated with tales of students such as 18-year-old Ben Scheffer, a public schoolboy from Brighton, who failed to get a single offer despite securing three A*s and three As last year.

Many such stories are "fairly mythical", claims Sir Steve Smith, vice-chancellor of the University of Exeter and former president of Universities UK. "They usually reflect bad advice by schools, where the individual has no insurance offer as backup." And under the White Paper's plans to ease restrictions on student numbers, "this will never happen" as institutions will be free to accept as many AAB students as they can attract, he says.

Smith adds that 2010's headline figure of 210,000 applicants was also much higher than the number finally left disappointed.

Almost half - 97,000 - declined offers or withdrew from the process, which left about 113,000 candidates unplaced from the 697,351 who applied last year. This compares with 79,000 who failed to gain places in 2009.

But the number of places on offer has increased over time. So the ratio of applicants to acceptances was 1.4:1 last year, only a slight change from the 1.3:1 ratio found in each of the previous five years.

Mary Curnock Cook, chief executive of Ucas, argues that the number of unsuccessful applicants has not grown massively despite popular belief.

She says a certain failure rate is inevitable. "It's always been about 100,000 people who just do not make the grade," she says. "These are the people on the margins of being qualified for ÖйúAƬ. We also get a lot of people who put in poor applications. If they do not (go to) any trouble with their applications, are they really making an effort to go to university?"

However, according to recent research published by the ÖйúAƬ Policy Institute, it is also true that many bright students may have missed out on places because their courses are not recognised by the Ucas tariff, the mechanism that many admissions staff use to compare different qualifications.

The tariff is under review: at present, only about one-third of Level 3 courses are included and the Hepi study found that nearly half of disappointed applicants did not have any tariff points.

"Some people say this is essentially something for universities to solve," says Smith. "Some say we should provide more access courses and assess people ourselves. But the debate needs to think about how many people are qualified to go to university, which is an issue linked to dropout rates."

'You need top grades to get into a top university'

It is commonly assumed that the elite Russell Group of research-intensive universities secures all the students with top grades. But official figures paint a different picture.

Research published in July by the ÖйúAƬ Funding Council for England shows that the destination of high-achieving A-level students is not so easy to predict. The data show that many AAB students attend institutions that do not traditionally top the university rankings. A study by the ÖйúAƬ Careers Services Unit found that Oxford Brookes, Northumbria, Glasgow Caledonian and Robert Gordon universities all sit in the "high-tariff" Ucas bracket, as does Queen Margaret University in Edinburgh.

Meanwhile, less than half the students at five Russell Group universities had AAB grades or better, with just 32 per cent of those studying at the University of Liverpool achieving those marks.

The location of high-achieving students "varies much more by subject than by institution", explains Anna Vignoles, professor of economics of education at the Institute of Education, University of London. "It's not just about the Ucas points you achieve - it's about subjects that you do well in.

"Private schools tend to choose the subjects favoured by universities, especially higher-ranking ones, which is why they get more children into the top universities."

That could explain the "stark inequalities" regularly highlighted by the Sutton Trust. A study by the education charity last month showed that students from state schools with the same grades as their independent-school peers were less likely to enter top universities. A handful of elite public schools dominate access to Oxbridge.

"People from wealthy backgrounds will get good advice on the A levels that universities want," says Smith.

However, disadvantaged schools trying to boost their results "may push students to do easier A levels that are not well regarded by higher institutions", he explains.

Sir David Watson, professor of ÖйúAƬ and principal of Green Templeton College, Oxford, adds that nearly all large ÖйúAƬ institutions are both "recruiting" and "selecting" for their courses. (In some subjects, demand is less strong.)

He also points out that many highly qualified students who choose to study at new universities know exactly what they are doing. "In some cases they will be choosing high-quality courses that make sense for them over institutional reputation," he says.

'A levels are getting easier'

The era of the "unfailable" exam is almost upon us, according to certain newspapers. Bemoaning the "prizes-for-all" culture of modern schooling, commentators say that the ever-increasing A-level pass rate is undeniable proof that exams are getting easier. The pass rate last summer rose for the 28th consecutive year, with 97.6 per cent of entries gaining an E or above.

And the proportion attaining higher grades has also risen. One in six students scored a hat-trick of As at A level in 2009, according to the Cambridge Assessment exam board, double the rate in 1996.

But does this mean that exams are easier? Curnock Cook says it is not that simple.

"The height of the hurdle is the same," she says. "But the government has invested billions of pounds in education, so you would expect more people to pass the exams. And when teachers and schools get used to a syllabus over time, they will get better at teaching it.

"People are also not put in for A levels unless they are going to pass them, which accounts for the high pass rate."

So does better teaching really account for such improvements? And does "teaching to the exam" produce students able to cope with the rigours of ÖйúAƬ?

Smith believes the picture is mixed. "It is undeniable that academics in science are doing more remedial catch-up work (with new students) than in the past. There are also gaps in the knowledge of arts and humanities students because there is less learning by rote. But there are, perhaps, better analysis skills. A lot of the work done by schools is much more useful than that done in my generation."

Michael Gove, the education secretary, thinks the answer is for universities to get more involved in the process of A-level design. He has promised to take the A-level syllabus and the question-setting process "out of the hands of bureaucrats and instead empower universities, exam boards, learned societies and bodies (such as) the Advisory Committee on Mathematics Education with the task of ensuring these qualifications are rigorous".

Some argue that the question of whether standards are the same as they were 30 years ago misses the point. What really matters in today's world is the international comparison: how our school-leavers compare with their overseas peers.

ÖйúAƬ

In any case, there have always been some examiners who have been critical of the work done by some students at A level.

Since the introduction of the exam in 1951, examiners' reports have been dotted with complaints about pupils' spelling, grammar and mathematical ability. A report on A-level maths in 1960, for example, despaired that, too often, candidates "had no understanding of the subject matter of most questions".

The Times Educational Supplement often published collections of exam howlers, including one from an English A-level paper in 1962 in which a pupil wrote: "The Friar preferred the company of baremaids."

'Predicted grades work against poorer students'

Each year, soon after A-level results day, there are often renewed calls for a revamp of the admissions system. Debate over the need for a post-qualification applications (PQA) system - which would allow students to apply to university armed with actual grades rather than teachers' predictions - has been rumbling on for years.

Seven years ago, the Schwartz report recommended PQA, arguing that assessment of students using actual grades made far more sense than the guesswork employed by teachers.

Criticism of the current arrangements is understandable given that about half of all predicted grades turn out to be wrong.

Bill Rammell, former Labour minister for ÖйúAƬ, claimed that "the existing system is least fair to the poorest students" when championing PQA six years ago.

However, advocates for reform are wrong to say that predicted grades discriminate against poorer students, whose teachers might be timidly underestimating their results.

In fact, says Smith, the inaccuracy benefits state-school pupils: "The most accurately predicted grades are As and independent schools are dealing with students who will be getting As." However, "poorer-performing schools ... might make predictions for more marginal students if they are aiming for a certain course. If they miss out by one grade, we might say it's not all gloom and doom and offer them a place.

"You could argue that the current system benefits pupils from poorer backgrounds."

Smith's argument is backed up by a report published by the Department for Business, Innovation and Skills in June about the accuracy of predicted grades. The BIS study, conducted by Ucas, found that in 2009, independent schools achieved the highest percentage of accurate grade predictions (64.7 per cent) - partly because 70 per cent of all predictions made were for A grades.

Only 30.7 per cent of grades were over-predicted at private schools, compared with 45 per cent at state schools.

Predictions for children whose parents had "higher managerial" positions were also more accurate, with 58.3 per cent achieving their forecast grades compared with 42.8 per cent of children from blue-collar backgrounds.

Over-predictions were more frequent among students from poorer backgrounds, with 49.5 per cent of grades overestimated compared with just 36.2 per cent among their richer counterparts.

However, Smith points out that about 90 per cent of predictions were not far out - plus or minus one grade - and errors for cumulative predictions (what the same student will achieve in different subjects) were even smaller.

Overall, nearly 52 per cent of grades were correctly predicted, with 41.7 per cent overestimated and 6.6 per cent underestimated.

Indeed, in 2005, Rammell was forced to issue an apology for making the claim that the system left the poorest students worst off. The coalition government has said it will await the outcome of a Ucas review of admissions processes before making any changes.

Politicians should be careful what they claim. In June, a press release from the office of Elizabeth Truss, Conservative MP for South West Norfolk, stated that "sciences, maths and languages at A level are increasingly the preserve of private and selective schools" while comprehensive-school students do media studies.

While it is true that, proportionately, pupils at private schools are more likely to take science, maths and languages at A level than those in comprehensives, the gap is not widening - it is narrowing.

The number of children from state-maintained schools taking core sciences at A level rose from 39,974 in 2005 to 43,388 in 2009, according to the Royal Society's State of the Nation report, published in February. Science take-up at academies and sixth-form colleges also rose: meanwhile, only 587 more independent-school children took science in 2009 compared with four years earlier (14,824 compared with 14,237).

It's all in the preparation

Neil Hopkins, principal of Peter Symonds College in Winchester, said the well-worn stories about "falling standards" were the most frustrating for teachers. "When Roger Bannister ran the four-minute mile in the 1950s, it was a big achievement," he says. "Nowadays it's fairly commonplace, but people don't question whether the mile has become shorter.

"The improvement is down to better training and preparation, and it's the same for A levels. My students work very hard, and it annoys me to see their achievements denigrated."

But 18 August will still be "an amazing day", Hopkins adds. "It's always so emotional - the students are full of adrenalin, anxiety, hormones and excitement. It's the moment when they see the outcome of so much work.

"Pupils are often weeping, but mostly with relief or because they can't believe they've actually achieved the marks predicted for them. If we were able to bottle the emotional energy on that day, we would be able to power the National Grid."

Grade A* in whooping: why media gentlemen prefer (private) blondes

Mockery of the images of pretty girls leaping for joy that inevitably accompany Fleet Street's coverage of A-level results has become almost as much of an annual ritual as the photographs themselves.

But this summer, Chris Cook, education correspondent for the Financial Times, decided to reveal exactly how English private schools compete to supply pictures of attractive young women to the national press.

"It is a well-worn observation that even respectable newspapers like to illustrate their coverage of A-level results day with pictures of delighted and pretty (usually blonde) girls," Cook wrote, before detailing an unsolicited voicemail he received from the press liaison at Badminton School in Bristol.

"Hi Chris...Just wanting to give you some details of some absolutely 'beyootiful' girls we've got here who are getting their A-level results tomorrow," the message went. "Some lovely stories...They're amazing girls."

Another school even invited Cook to an end-of-year sports event - an invitation he declined.

One blog, It's Sexy A-levels!, has formulated rules for the perfect picture, with "blonde" "twins", "going to Oxbridge", "leaping for joy" and "holding aloft results" among the key ingredients.

Photos featuring synchronised leaping, ecstatic hugging and whooping earn bonus points.

Why A* grades at A level are not always the best guide to picking top students

Faced with complaints from elite universities that the growing number of A-level A grades had made it harder to select the best applicants, the answer seemed simple: the A*.

The higher grade was introduced last year, with students achieving an A* in 8 per cent of exams - the same proportion that gained an A grade in 1965.

But several academics have suggested that the A* may promise a level of precision it cannot deliver.

Sir David Watson, professor of ÖйúAƬ and principal of Green Templeton College, Oxford, explains: "If you are going to divide the A grade into A and A*, a single mark can make a huge difference to the fate of the candidate according to where it falls...

"To achieve an A*, a candidate has to get 320 points across the four units, including at least 180 points at A2. So if Candidate 1 gains 60 on each of her two AS units and 100 on each A2, she scores 320 and an A*. Candidate 2 may get 59 on one AS, 60 on the other and 100 on each A2, achieving 319 and missing an A* by one point.

"But Candidate 3 may score 100 on each of her AS units, 90 on one A2 and 89 on the other, a total of 379 - but only receive an A. That person scored a whopping 59 more points than Candidate 1 and 60 more than Candidate 2, but lost out because of where their points fell."

Dylan Wiliam, emeritus professor of educational assessment at the Institute of Education, also believes the grade is statistically more prone to errors.

Fewer marks in exams will relate to an A* than they will to lower grades, he says, so any marking anomalies will have a bigger effect.

Robert Coe, professor of education at Durham University, argues that the A* does not recognise excellence or originality. Instead, it rewards the avoidance of careless mistakes rather than the exhibition of flair for a subject, playing into the hands of those best coached in exam techniques.

Official statistics from the Joint Council for Qualifications show that 30 per cent of all A* grades in 2010 came from private-school candidates, despite the fact that these students made up just 14 per cent of overall entries.

Meanwhile, in September 2010 the Times Educational Supplement reported that exam boards had downgraded more than 10 per cent of A-level results that summer in order to avoid a glut of A*s, prompting further questions about fairness.

A spokesman for the exam regulator Ofqual says: "A* grades are awarded to learners who consistently perform well throughout their A-level units and outstandingly on the A2 units.

"It is an important piece of information to help admissions tutors make decisions about candidates, but it is not the only information they will be considering.

"We are monitoring, as we did last year, the awarding of the A* grade to make sure it is fair and consistent. All candidates who receive an A* grade will do so because of exceptional performance."

ÖйúAƬ

A spokesman for the Department for Education adds: "A* represents genuine top-level attainment. The top universities have long wanted to differentiate between the best pupils."