Welsh Language Comparisons
Welsh Local Labour Force Survey and Census data
Both the Census and the Welsh Local Labour Force Survey (WLLFS) asked respondents in Wales if they could understand, speak, read or write Welsh. The Census gave lower estimates for these skills.
The Office for National Statistics (ONS) has undertaken to examine the data and give an explanation for these differences so that users can make an informed decision about which estimate to use. This work is currently being carried out and a full paper will be published in summer 2003. However, we can give a broad overview of the principal reasons for these differences.
The concept of "understanding Welsh" is poorly defined and could vary from understanding a few words to being fully fluent. Respondents will have applied their own definitions. These definitions will have been heavily influenced by the context of the question. The same is also true of speaking, reading and writing Welsh.
Respondents are more likely to say they understand, speak, read or write Welsh if they have just given their national identity as Welsh. The WLLFS has a question on national identity immediately before the Welsh language questions whereas the Census did not.
The 2001 Census form was completed by respondents, or by a household member on their behalf, usually without reference to an enumerator or any other person employed by ONS.
Every effort was made to ensure that the question was easy to understand and was "user friendly" but, as with all self-completion forms, data quality is unlikely to be as high as in a face-to-face interview. In particular, since most questions on the Census asked for a single answer, respondents may not have realised that they should tick all the answers that applied.
printer-friendly
Census homepage