Scotland the process was less straightforward as there was less consistency between the way they described some concepts and the same concepts in England, Wales and Northern Ireland. to match their data to the working model resulting from processing the ONS and NISRA. There was a need to interpret the meaning of the data and so we elected to err on the side of caution. If after looking at other examples and actual data values we couldn’t be sure it was the same concept, we then added new codes to the model. Scotland did however produce some UK harmonised data, which was easier to match in. An example of this would be Country of Birth.
Of course, we would have had the same issues vietnam rcs data if we had processed the Scottish data first and had been matching the ONS and NISRA into that model. The very nature of bringing data together that has been produced by different agencies (who all have slightly different user requirements) is fraught with these kinds of difficulties.
Another complication to processing their data was that data were released in batches, which meant that if in a later release a new variant of a concept cropped up, or there was additional information in the footnotes for an existing concept released in a previous batch, it meant we had to go back and reprocess all of the instances of that concept. Then there was the possibility that in a later batch of data, new information may arise that caused a re-evaluation of these concepts. Scotland produced about 15 batches of data.