Funferal

2002 census – the answer

Have I mentioned my admiration for the Irish CSO? Yesterday I posted a question about the date of the last Irish census, and simultaneously sent a query to the CSO’s standard email address. When I got up this morning – by 3pm Irish time – I had a response.
First, the delay was due to the outbreak of foot-and-mouth in 2001. Irish readers will remember the extreme measures taken to prevent the disease spreading, and one response was to delay the census – since census gatherers would have to traipse from dwelling to dwelling, potentially spreading F&M from farm to farm.
Such a simple reason, yet one that had, for some reason, escaped me.
Second, I asked about the impact of this on the utility of the information, given that the five year cycle is interrupted. Here’s what the CSO’s senior statistician had to say:

Clearly, holding censuses at five yearly intervals enables cohort analysis to be done on the basis of five year age group data. The fact that Census 2002 is six years after the previous census implies that single year of age data has to be used for this purpose. It is of interest to note that the five year sequence was also broken during the 1970’s when we held censuses in 1971 and 1979.