The initial product of a MCA is a large matrix of scores of each option against each of the criteria (see above). But this often provides too much data to be intelligible. The key issues that determine the choice need to be ‘pulled out’ from this data. Of the two approaches to choice, ‘focusing in’ and ‘sieving out’, sieving out removes the irrelevant from further consideration and is to be preferred. In making a choice, we finally then only have to select a single option as being preferable to all others. The rank order of preference between the remaining options is then irrelevant. So, the first principle is to get rid of the irrelevant in order to identify the important for attention.