6. Is the data plausible?

Another check that can be made to see if the data is fit for purpose is to check if it is plausible. For example, asking a group of people about their height and one answer obtained is 9 foot.

This is quite an unlikely height i.e. the data is not plausible. So this would flag that further checks need to be made to trust the answer.

Whenever data is wildly out of the norm, it should be questioned for 'plausibility'.

Plausible data is improved if:

You trust the source of data

For example, if you wanted a data set of rainfall in the UK, the most reliable source would be directly from the Met Office. Copying it off someone's blog site is less plausible.

The source should have a reputation of being accurate and reliable.

The data is reasonable

For example, you want to store the likely unemployment figure for next year. No one really knows the answer but one way is to ask a number of experts their opinion, then a 'consensus' is reached where most experts agree on a reasonable number.

How was the data collected?

The method of data is crucial in working out if the data is plausible. For example, a data logger measuring daily temperature in the UK suddenly start recording 150 F or -50 C. This is clearly not plausible.