MAIN LANGUAGE OTHER THAN ENGLISH SPOKEN AT HOME
This topic provides information about the main language other than English spoken at home, if any. Information on language use is important for a wide range of policies at the national, state and territory, and local levels.
A question on language was first included in 1921, and asked whether the person could read and write. In 1933, the question asked whether a person could read and write in a foreign language if they were unable to read and write English. A question on language was not included again until the 1976 Census, when a question was included on all languages regularly used. In 1981 a double-barrelled question was included, the first part of which was about whether the person spoke a language other than English at home. If the person answered yes, he or she was was sequenced to the second part of the question asking about proficiency in spoken English. In every Census since 1986 a question on language spoken at home has been included, with those who answered yes sequenced to a separate question on proficiency in spoken English. From 1991, a list of response options has been provided for the question, reflecting the most common languages reported in the previous Census. Respondents are able to provide their language in a text box if it is not included in the list of response categories.
Despite not being one of the most commonly reported languages in the preceding Census, Mandarin was included in the list of response options until the 2006 Census to cover the main Chinese dialects, and to minimise the possibility of people who speak the Mandarin language opting for Cantonese in the belief that it was the only Chinese dialect available. For the 2016 Census, Mandarin will again be included in the list due to the frequency of responses in the 2011 Census.
In the lead-up to the 2006 Census, the ABS considered changing the language question to collect data on English and Other only, with Other being unspecified. An investigation on the modelling of language data was undertaken, using responses from related ethnicity questions from both the 1996 and 2001 Censuses. The results showed sufficient quality outcomes for a range of languages at the Statistical Local Area level. However, the investigation also showed that some areas of the Language Classification would experience high levels of either over, or under, estimation of populations if modelling was relied upon for the derivation of language data. This was especially true for Indigenous languages. Consequently, the format of the 2001 Census question has been retained in subsequent Censuses.
The following question is from the paper 2016 Census Household Form.