IF YOU have been reading the business pages and seen reports about newspaper readership and newspaper circulation on the past two Fridays, you could be forgiven for wondering what is the difference between the two.
And, if you read the report this week in The Australian Financial Review, about the problem of "mouse fraud" in measuring the popularity of internet sites, you could be forgiven for worrying about the vaunted precision of new technology.
All three reports are about media companies' need to be able to deliver to advertisers "the people formerly known as the audience", as Jay Rosen, an American media professor who runs a weblog, PressThink, puts it.
As the tectonic plates of the mediascape shift, the task of connecting audience with advertisers is far less straightforward than it has been for the past 150 years or so since the rise of the modern commercial newspaper.
The Audit Bureau of Circulation has been counting paid sales of Australian newspapers since 1932. Since the 1970s, the Roy Morgan Research company has measured readership on the premise that newspapers and magazines are read by more people than the purchaser.
Morgan researchers survey people around Australia aged 14 and over about their newspaper and magazine reading habits and attempt to calculate the number of readers per issue.
Over time they have developed two rules: first, there are more readers per issue of a magazine because magazines have a longer shelf life than the daily newspaper, and second, specialist publications have fewer readers per issue than general interest publications because they are more narrowly aimed at particular readers.
Broadly speaking, magazines are read by three to four people per issue while newspapers are read by between two and three people.
The Morgan readership figures, which were published the week before last, are calculated by multiplying the circulation figures by the number of readers per issue.
Newspaper sales slowly declined for many years, especially as television rose to displace newspapers as the most popular mass medium.
In the past few years, the internet has become another mass medium to challenge the primacy of newspapers and, for that matter, television.
For many years, according to Ivor Ries, a media analyst with E. C. & L. Baillieu, newspaper companies inflated their circulation figures by counting copies that were given away or sold for a token amount to airlines, hotels, schools and cafes.
Similarly, until the 1990s the industry-wide decline of newspaper sales was not publicised in the media.
That has changed as the internet's popularity has tightened competition for advertising dollars, and as the technology surrounding the internet has enabled media companies — and advertisers — to measure more precisely who reads what online, and for how long, through figures provided monthly by Nielsen Net Ratings.
PricewaterhouseCoopers, in its recent comprehensive forecast report Australian Entertainment and Media Outlook 2006-2010, said: "Issues with the accuracy and frequency of newspaper circulation and readership information have dogged the industry for years."
Calls to reform the system have intensified in recent years because, says the report's principal author, Matthew Liebmann, the internet can provide "very robust and granular information".
The ABC circulation figures that were reported in the press last Friday were the last to be done on a six-monthly audit. From July newspaper and magazine sales are being audited quarterly.
New guidelines issued by the ABC also require newspaper and magazine publishers to specify the number of average net paid sales they made in four categories that in the past were associated with giveaways.
These categories include sales to hotels, airlines, schools, through universities, and at special events such as the AFL grand final or the Melbourne Writers' Festival.
Such reforms should strengthen the veracity of circulation figures. Meanwhile some newspaper and magazine publishers, notably News Ltd, have contested the accuracy of the Morgan readership figures.
The internet ratings figures are not beyond reproach, though, as the mouse or click fraud phenomenon illustrates. Mouse fraud is where a person clicks repeatedly on their computer to ramp up the number of "hits" for a website.
Potentially, it threatens the "granular" nature of the information provided by websites to Nielsen Net Ratings, but media analyst Ries says savvy websites overseas either have included an allowance for click fraud in their ratings or have offered advertisers a discount for any proven click fraud.
The trend overseas, especially in the United States, is for websites to develop technologies that enable them to provide what Ries calls "contextual" advertising.
For instance, when, as a subscriber to the online edition of The Wall Street Journal, he visits the site, he sees an advertisement inviting him to apply for a credit card for an Australian bank.
In signing him up, the website gathers information about his background, income, address, occupation and so on that enables them to provide targeted advertising.
In Australia, Nielsen Net Ratings is heading in this direction; it provides its subscribers with demographic information about website browsers, such as gender, age and income levels, through sample interviews.
Understanding the background and rationale of all these different measurement methods enables you to see the picture emerging from the latest round of figures, and it is a brighter than expected picture for the media.
It is this: the rise of the internet as a medium is not killing newspapers and magazines for advertisers, or for journalists who learn to avoid duplicating material in an already crowded information galaxy and to exploit the particular characteristics of each medium.
