Grooming the forms

When the boxes of IHFs from a particular Collection District (CD) arrived at the CMU, they should have had the following accompanying documentation:[1]

On the last of these, which the CFO was expected to complete before sending the forms to the CMU, the CFO was required to crosscheck, among other things, that:

In practice, many of these CFO checklists were not filled in, so the CMU could not know whether or not the CFO had carried out the checks asked for, and very little attempt had been made to update the DICD forms. In many cases, the IDCs were also missing.[2]

The community checklist for the CMU had two sections. One—theoretically to be completed while the CFO was still in the field—asked the CMU checker to make sure that the CFO had recorded and reported the counts for people, dwellings and unoccupied dwellings. At the CMU, these counts were then checked against available estimates and the CFO notified of discrepancies for follow-up. In practice, this happened after the forms had been returned to the CMU and the CFO was either no longer in the field or was in a different CD, necessitating in some cases a return to particular CDs, or a follow-up by phone or fax.

The second part of the CMU checklist involved inspecting the forms to ascertain whether the CFO had carried out all the checks that they had been asked to do on the CFO checklist. In addition, the checkers at the CMU were asked specifically to do the following:

All of these checks—and most of the checks that the CFOs were asked to undertake—were concerned ultimately with ensuring the consistency of the ‘head count’ aspect of the census and ensuring that as far as possible basic information about residence and Indigenous status had been recorded. All things being equal, they were designed to ensure that the forms sent to the DPC were internally consistent with respect to the number, sex ratio, usual residence and Indigenous status of the population. In terms of the accuracy of the count, the checks were initially more general in nature: the CMU was concerned if the population count for a particular CD or the number of dwellings for which forms existed was very different from the recently acquired CHINS data and/or the count at the 2001 Census. There were several cases in which CFOs were sent out again to investigate the reasons for discrepancies.

There were several issues where the checkers were told not to try to ‘fix’ discrepancies, for example, anomalies in ages and/or in the data on relationships between the members of the household. They were told that they had insufficient information to make changes and that the DPC had rules and procedures that applied in such cases. Again, this was an example of the application of prior knowledge about the context into which the forms were being sent.