Whoever chose the title for the annual volume of educational statistics from the Organisation for Economic Co-operation and Development either had a sense of humour or a bad translator.
As the 1996 tome has 396 pages and is accompanied by a 75-page analysis, it isn't immediately obvious why the series should be called Education at a Glance.
If you are a researcher or educational "trainspotter" who is keen to know the percentage of 25 to 64-year-old Spanish graduates who were unemployed in 1994, or the proportion of Mexican infants in private nurseries, these are the volumes for you.
But the primary function of these statistics is to give the OECD's member governments some insight into how their educational inputs and outcomes compare with others.
Some nations, such as the United States, have always heeded education statistics.Others, such as Germany, have refused to let their policies be affected by such comparative tables. This is arguably the more sensible strategy because even the OECD admits that "the science of understanding and interpreting international education indicators is still in its infancy".
Furthermore, statistical tables cannot answer the one conundrum that frustrates all governments: "What can we do to improve our educational performance relative to other countries?" It also has to be recognised that some of the statistics that the OECD published last week should be consumed with a sprinkling of scepticism. Others are simply wrong. As the Guardian pointed out this week, the Education at a Glance analysis wrongly reports that the salary of experienced UK primary teachers improved by only 2.5 per cent in real terms between 1985 and 1993. The real figure, given in another of the booklet's tables, was 40 per cent.
But the OECD statisticians should not be derided for that bloomer. As the data are derived not only from the OECD's 27 member governments but myriad other sources, some errors will always creep in. There are, however, other problems associated with these indicators that are harder to rectify, and the OECD acknowledges this.
The educational outcomes of different countries cannot be compared in the same way as their Gross Domestic Products. A-levels are not the precise equivalent of either the French Baccalaureat or the German Abitur, and matching vocational qualifications is mind-bogglingly complex. Even university qualifications do not have the same "exchange rate" because a first degree can be obtained in three years in England - two years quicker than in Germany.
The time lag associated with all research of this kind also means that Education at a Glance provides us with a better picture of where we have just been than where we are now. The "new" OECD statistics on overall education spending, class size and teachers' salaries take us only up to 1993-94, which probably satisfies historians but invites scorn from journalists.
Education at a Glance also cannot convey the full picture on nations' educational investment because some governments have been unable to quantify private education spending. The Koreans, for example, are known to invest about 2 per cent of their GDP in private schooling - a colossal sum that is largely spent on crammers - but this has not been officially reported.
Readers of these volumes should also be aware that as the OECD is funded by its member governments it has to consider what it would be politic to publish. There is therefore no data on drop-out rates from higher education, although statistical sleuths could doubtless produce their own reasonably accurate estimates.
More worryingly, although some countries are prepared to take criticism on the chin it is not unknown for a government to insist that unfavourable data should be expunged from the official record. Last year's seven-country adult literacy survey should have covered eight countries, but France pulled out shortly before the publication of last December's report.
That came as no surprise to some adult educationists because France has traditionally been loath to recognise that it has an illiteracy problem. However, such incidents are potentially much more damaging to the credibility of OECD statistical publications than a few mistakes attributable to human error. No doubt the OECD realises this, but given its dependency on member governments it is difficult to see how this can be prevented in future.