The ‘need to know’ criterion should also play a key role in
helping you decide what level of reportage is appropriate, the
right degree of detail. Suppose that I want to quote a UK labour
market number at one point, and official sources give the num-
ber of unemployed people as 1,215,689. The usual academic pro-
cedure would be to just quote this number in full, unmodified in
any way. But a number of issues arise. Do readers really need to
know this exact number? Do they care whether the number is
exact to the nearest one person, or the nearest ten, or the
nearest hundred or thousand? In the context of your argu-
ment would they lose any significant information if the
number was expressed as 1.21 million unemployed, or even
1.2 million?
Some university people will immediately bristle here at the
idea that as authors they should fillet out or reduce the level of
detail conveyed by their text. Their view might be that it is not
their job to ‘pander’ to lazy readers, or to make things easy for
people. In the social sciences, some critics suggest that there are
many academics who suffer from ‘physics envy’, a desire to ape
practices in the physical sciences in pursuit of enhanced aca-
demic prestige. Whatever the truth of such claims, there are cer-
tainly many people who seem to regard the citation of complex
numbers and multiple decimal points as essential talismans of
systematic scientific endeavour. Not for them the production of
‘easy’ text, but instead an emphasis on precise accuracy in
reportage at all times. But consider for a moment the ‘scientific’
implications of reporting 1,215,689 unemployed people.
Including such a precise number in your text suggests that you
believe the accuracy of government counting systems is plus or
minus 1. Quoting this number in full also means that you are
confident the real figure is not 1,215,685 or 1,215,691 people,
but exactly 1,215,689. In fact it is highly unlikely that the offi-
cial statistics are that accurate. A genuinely scientific approach
would be to report information only correct to the number of
digits where we can have reasonable confidence in the data.
Worse examples of completely bogus ‘scientism’ in the
handling of many numbers occur in many PhD theses. It is
common to see students making elementary mistakes like the
following. Suppose that in a national survey of 1021 respon-
dents, 579 people report that they have tried surfing the
HANDLING ATTENTION POINTS◆ 161