Average Issue Readership – There Is Something Wrong
Last night’s heated debate on the accuracies of the NRS figures resulted in no real answer to the questions raised. After the award winning paper from Neil Shepard Smith, the NRS made a spirited response. The overall feeling from the floor was one of confusion. There were calls for the NRS to accept it has problems and to address them rather than deny them.
Neil Shepard Smith’s paper, “Average Issue Readership-There is Someting Wrong” proposes that the average issue readership figures cannot possibly be correct; they are artificially high.
Neil attributes the problem to the “recency” method, where respondents are asked when they last saw a copy of a publication; if they claim to have done so in the last publishing interval, they are included in the average issue readership. This method would work if the last reading event was the first time the respondent had seen the magazine. If the magazine is read again at any point outside the issue-period in question then it is counted again. This “replication” can seriously inflate the average issue readership. Neil maintains that the maximum possible number of readers per copy can be mathematically calculated using source of copy data. By this method NRS data gives impossible figures.
As alternatives to the current NRS calculation methods Neil suggests the FRIPI method, or First Reading in the last Publishing Interval. The first reading question follows the establishment of any reading withinn that issue period. This system is used in South Africa. Neil also advocated the use of an estimate of reading days, which could be provided by developing the MPX approach.
Roger Beeson began the NRS defence by agreeing that there is something wrong, but “with Neils’ arguments”. The NRS defence was accompanied by cartoons of Neil as a heretic, with NRS members in various stages of carrying out an exorcism.
Chief exorcist was Michael Brown, Technical Consultant to the NRS. He began by telling us that “Neil Shephard Smith distorts the truth”. Michael pointed out that Neil had ommitted any reference to parallel readership, which counterbalances any exageration in figures caused by replicating. If more than one issue is recorded during the publishing period this is lost in the NRS data.
The second theme of the NRS defence was that Neil’s calculations which show that the NRS figures must be wrong depended on Source of Copy data. This, said Michael is taking source of copy data too literally; as it stands, the data is only meant to provide primary readership data.
The debate then transferred back to Neil Shepherd Smith to reply; in response to the attack that he did not refer to parallel readership, he called this a red herring, adding that there were many aspects of the NRS that he did not refer to, which were irrelevant to the argument.
As to the response that source of copy should not be used in this way, Neil pointed out that the NRS was admitting that their own figures were not accurate enough, and was therefore a useless argument. Just when all arguments seemed to be going round in circles the debate was opened to the floor.
One thing which seemed to worry the members greatly was, as John Billett commented, the fact that the NRS itself said that Source of Copy data was not accurate enough to use. Harold Lind echoed these worries, and suggested that the NRS turned to provide data on how many ads are seen rather than readers per copy.
There was however, some support for the NRS from Guy Constantine. He referred to a MRG sub-committee formed in 1988 which had found that parallel reading could feasibly off-set the replication question, as NRS had said. They had found that copies in public places could generate high readership, as could friends with like interest sharing copies. Alan Smith pointed out that replicated readership should not in fact be rejected so damningly, because to advertisers it can be very valuable, as it offers more chances to see ads. Alan Copage, chairman, asked for a show of hands to answer how many members were satisfied by the NRS defence. Not many said “totally”; an equally small amount said “not at all” with the vast majority deciding on “partially”. As Alan said, “this debate could run and run”.
To close the evening, Ray Morgan, the new chairman of the NRS, told members of the forthcoming data on newspaper sections. Topic of Interest data on both newspapers and magazines is also on the agenda.