Page 77 - Living Room WarsDesprately Seeking the Audience Rethinking Media Audiences for a Postmodern World

The people meter 'solution' 65
stressing the methodological imperfections of the people meter. For example, NBC’s vice
president of research William Rubens complained in a speech that the changeover to
people meters ‘has brought home to me how far away the research business has drifted
from sound methodology’. He went as far as condemning the fact that ‘market pressures
forced [Nielsen] to introduce a ratings service before its time’, and argued that ‘critics of
the old system should be just as skeptical about people meters, for if we don’t have the
best media measurement money can buy, we will wind up with inaccurate ratings’
(Broadcasting 21 March 1988:27). CBS’s head of research David Poltrack (1988), too,
asserted that ‘the industry’s standard of living’ has ‘lowered’ since the introduction of
people meters. Thus, soon after the people meter’s official inauguration, ABC, CBS and
NBC took the initiative, together with the National Association of Broadcasters, in
requesting Nielsen to co-operate with an independent evaluation of the people meter
methodology (Broadcasting 21 March 1988).
In formulating criticisms of the people meter, strikingly similar arguments are
reiterated as those once levelled against the old paper-and-pencil diary technique: they
involve too much subjectivity. The term ‘people meter’, it is contended, is a misleading
one because a meter is supposed to measure automatically, and should not require the
cooperation of those being measured. Rubens has sneeringly dubbed the people meter an
‘electronic diary’, requiring ‘people pushing buttons instead of pencils’ (in Beville
1986a:78). And indeed, it is this button-pushing chore that has become a focal concern
among the critics. A general discourse of distrust has developed around it: how can one
place reliance on people’s veracity in handling the pushbutton gadget? Mal Beville, the
doyen of ratings specialists, has cited many problems: ‘Unlike the diary, where recording
errors can be corrected after the fact, the people meter requires perfect instantaneous
button-pushing performance. No second chance is available to correct a mistake.’
(1986a:78) And another professional observer wondered: ‘Will the families in the sample
really take the trouble? Will they always press the buttons as they begin watching? Will
they always remember to press their buttons when they leave the room—as when the
telephone rings, or the baby cries?’ (Baker 1986:95).
In an environment where so much depends on the figures, these worries are perfectly
understandable and logical. Therefore, from the very beginning the ratings firms
themselves have done their best to determine what they call the ‘compliance rate’ to the
device, to see to what extent people can be motivated to co-operate with the system
3
(Gardner 1984). But doubts have not calmed down. Support for that doubt was delivered
by a test with Nielsen’s initial sample of 1,000 households, which according to Poltrack
showed that ‘half of the people aren’t cooperating and only 75 per cent (of those) offer
good information each day’ (in Donlon 1987). Such low ‘compliance rates’, the network
researchers were eager to emphasize, will inevitably lead to all sorts of bias in sampling
and in the resulting data. For example, they were extremely taken aback by test results
that indicated that the technology seemed to be more acceptable to larger families, to
those with higher income and education levels, and to people familiar with ‘high tech’
items such as personal computers and VCRs, and less in blue collar families and among
older people in rural areas. Furthermore, children, teenagers and women especially
seemed to be less ‘reliable’ in their push-button behaviour. And lastly, they were
sceptical about whether sample families will not get tired of pushing buttons day-in, day-
out, for the whole period that they are part of the sample (which can be years), every time