Running alongside this column is the industry's most comprehensive and, one likes to think, accurate chart of the top-grossing films during the past weekend in the international marketplace.

It should be pointed out that even such an established chart faces problems that are intrinsic in putting together a compilation of the hit movies playing on screens from Ankara to Zagreb. The most obvious hiccup is monetary conversion rates.

Leaving aside territories where information might be suspect, all one has to do is to look at the top grosser in the UK last weekend.

Crunching the numbers

According to one source it was Mr Bean's Holiday, with weekend revenues of $2.89m; another reported the figure at $2.99m; and a third calculated it at $3.03m.

In the rarified climes of box-office reporting, one hopes to work with a margin of error of +/-2% but taking the high and low figures in this sampling reveals a 4.8% spread.

It gets more complicated. Yet another source posted a Bean weekend of $3.02m. However, it ranked the comedy second to Wild Hogs with a gross of $3.34m. What this renegade report did (and others did not) was to include pre-weekend previews while the other charts reflected that revenue in the picture's cumulative total.

It is a confusing picture.

There is nothing essentially wrong in either including or separating preview grosses in a weekend report. The problem is that in none of these four instances is the practice consistent. Any one of the suppliers will include or exclude previews in their reporting and this is due in part to the human factor.

Open to interpretation

Different people compile data for the same companies and depending on the person - or instructions from a superior - that part of the box office can be interpreted in several different ways from week to week.

Screen International's policy is to report preview revenue at the time it occurs. Admittedly sometimes distributors are not forthright about such things.

During the prior two weekends, our practice resulted in 300 ranking first on the March 30 weekend and Mr Bean moving into the top slot the following weekend. However, other press reports reversed the situation and most reporters on those titles do not have a mechanism to check for accuracy.

Another anomaly is the simple definition of 'weekend'. Americans tend to view it as a three-day event, except during holiday periods that can extend the span to five days. In regular circumstances, however, Japan reports a two-day weekend, Mexico reflects the American notion, Australia takes four days and France spans five.

There is no question that people's best efforts are involved in producing an international weekend box-office chart and we are justly proud of the one on this page.

This chart was the first and all others have to be measured by the standard that has been set. And it is absolutely accurate... within the fuzzy parameters it measures.

[s19] Feedback: