Subsequent development of the enumeration strategy in remote areas has essentially modified and extended the practices set in place in the Northern Territory, Western Australia and South Australia by 1981. Thus, the key operational features that set the remote area IES apart from the general census remain the use of interviewers and the administration of different forms with a modified format. While a systematic approach to interviewing and form-filling is outlined and encouraged in pre-census training, an ethos of local control and flexibility regarding just how the necessary information is to be acquired is also instilled. This can create additional idiosyncractic departures from general census practice.

The basic procedural structure of the IES is still as illustrated in Fig. 1.1, which shows the procedure for the 1991 Census. Using a Dwelling Check List (equivalent to the Community list in Fig. 1.1; see Appendix A for the 2001 Dwelling Check List), the first task of community coordinators (CCs) is to compile a comprehensive pre-enumeration listing of all dwellings in their community and to label each of these with a family name. Ideally, an estimate of the overall numbers of males and females assumed to be resident in each dwelling and available for interview should also be indicated. This process commences as soon as CFOs can organise and train an appropriate individual in each centre to do the job. Cross-checking of community lists with housing and other administrative lists is also encouraged. The method of compiling lists of dwellings and families may also vary according to who is involved, although this is generally done with assistance from council officials and other community record keepers.

Once constructed, the Dwelling Check Lists provide the basis for completing Special Indigenous Household Forms (SIHFs; see Appendix B for the 2001 SIHF), with one form for each household group (step 2; the SIHF corresponds to the Remote area household form of Fig. 1.1). In turn, Special Indigenous Personal Forms (SIPFs; see Appendix C for the 2001 SIPF) are completed (step 3; the SIPF corresponds to the Remote area personal form in Fig. 1.1) to record the personal details of each household member identified in step 2. The last two steps are, in theory at least, carried out by locally recruited interviewers in an interview with an identified household head—in effect an adult person with authority to provide information on behalf of all household members. In practice, the information may be supplemented from second-hand sources such as key informants and administrative records.

The process of administering each of these procedural steps inevitably involves a time lag with potential (and ultimately unknown) consequences for the accuracy of the count in terms of overlaps and omissions, especially in view of the high level of intra-regional population mobility in many areas. In recognition of this, fairly drastic measures are sometimes devised. In Yuendumu, for example, census day is brought forward by two weeks in those census years in which the real date coincides with the Yuendumu Sports Festival. This event draws large numbers of people into Yuendumu from all over central Australia (Young & Doohan 1989: 192–7).

- 8 -