Unit 4 sec 2.4
Spurious precision
21 November 2015
15:25
The values, examples of spurious precision. In the 1st
case, the data value seems to be the result of the person’s age being given as
29 years and 11 months, and 11 months being 11/12= 0.916666 years. It is likely
that this value is entered as 29+11/12, which would automatically be displayed
in its decimal form. In the 2nd case, the spurious precision has
arisen by conversion, 2 kg, of data measured in different units (pounds).
Another way that spurious
precision can arise is when figures are quoted to a greater number of
significant figures than is warranted in the context.
The accused was a council employee who regularly stole a
portion of the money she was coming from the fees paid into machines by
motorists in car parks. Did you merely keep a careful record of all the money
taken and add it all up accurately.
Generally speaking,
it is customary, when analysing data gathered by others, to assume that the
claimed precision is justified unless there is definite evidence to the contrary.
Displaying data it is frequently unnecessary to retain their full procession. A
key principle here is that displayed data should be just precise enough to
reveal the key features - offering the reader an answer containing too many
significant figures can easily obscure these patterns in mass of numbers. If
you are collecting primary data, you should bear in mind the kinds of
difficulty with data discussed above in the case of secondary data, ideally you
should be able to go back to the original data collector in check with them.
Any suspicions you have about the data.
No comments:
Post a Comment