This section on the training received by the CFOs in Darwin must be read in context: the training is being evaluated with the benefit of hindsight. I will tend to focus on those aspects of the training that did not prepare the CFOs adequately for their task, in order to provide useful feedback to inform the training process for 2011. I want to stress at the outset that there were many positive aspects to the training, not least the amount of time and effort that went into preparing the training materials and the openness of the CMU staff to comments and suggestions from the CFOs and Assistant CFOs.
None of the CMU staff involved in managing the conduct of the IES in 2006 had been a part of the CMU in 2001. They were therefore heavily reliant on the advice and training materials provided by the central office, and on gleaning information from those people at the CMU who had experience of the count in 2001.
It is my impression that the practical knowledge and experience gained on the ground in successive censuses is fed back to Canberra in the form of reports from the CFOs and CMU staff, and it then informs central planning for the next census. It appeared to me, however, that very little of that practical knowledge and experience had been preserved at the local level from 2001. In other words, there appeared to be little institutional memory at the local level for the CFOs to draw on. The information that comes from the centre is informed by past experience, but the reasoning behind decisions that have been made is opaque to those who then have to use that information locally. They are presented in the training with a plethora of procedures and forms, with little sense of the real-life contexts in which they are going to be using them.
For example, it would have been invaluable to the CFO in charge of the eastern Arnhem Land region (see Chapter 4) to have access to the journal—or an edited version of the journal—of the CFO who had been in charge of that same area in 2001. The CFO for 2006 would then have had the benefit of the local knowledge gained by his predecessor and also the opportunity to learn from his mistakes and build on his successes. This information, together with updated forms for each community from the DICD, would give the CFOs a much better ‘feel’ for the real-life contexts in which they would be operating, and would give them a context in which to understand the procedural details on which the training inevitably focused in an exercise as complex as the census.
In the first part of the training session for the census, the CMU staff asked for feedback from the CFOs on the CHINS training that they had received. The CFOs were quite critical, feeling that the training period had been too compressed and that there had been too much delivery of information in a lecture format and not enough practical scenarios. They felt there had been too much emphasis on procedural matters and not enough on the content of the survey instrument. For example, the ABS definition of ‘temporary dwelling’ differs from the one used by other agencies, and this caused some problems with IHOs because funding is dependent on such definitions.
The general tenor of these comments also applied to the census training itself. It too was very concentrated and the information was presented in a highly condensed way, with little opportunity for the CFOs to practice what they were being taught. Rather than concentrating the training into a solid block, it would have been more effective to have half-day or one-day sessions spread over several weeks, concentrating on one aspect of the exercise at a time. One of the regional managers delivering the training commented to me that if people had read all the documentation they had been provided with before the training, not so much time would have been spent on unimportant questions. The trouble was that there was an enormous amount of documentation. If the training had been broken up into smaller blocks it would have been possible to ask the CFOs to review the documentation for that topic alone before the training session, and then re-enforce and amplify the written materials. This would have been a more effective way of ensuring that they were familiar with those materials.
Breaking the training into blocks would also allow time for practical sessions in which the CFOs would have the opportunity to consolidate their understanding. Scenarios would have been particularly beneficial in the session on providing training to the CCs and CIs. Most of the CFOs were not trained ‘trainers’, as they commented themselves. Giving them practical experience—by asking them to practice delivering training to one another in a mock tutorial—would have been an effective way of building their practical experience, supplementing the information that they were given on different ‘learning styles’.[2] This information, although interesting in itself, was not presented in such a way as to feed into the practical situation of a training session.
The focus throughout the training was on processes and administrative procedures. This was understandable, particularly in view of the complexities involved in some procedural matters—for example, the payment system for CCs and CIs.[3] One overriding concern—that the count should be as accurate as possible and be seen to be accurate—dominated the discourse. The focus on this concern had two effects. The emphasis on the ‘head count’ aspect of the census meant that the details of the questions on the Interviewer Houshold Form (IHF) received relatively little emphasis. And the focus on checking the count against other data—such as the 2001 Census, the 2006 CHINS and any administrative data that the CFOs could access—in order to assess the accuracy of the count led to a de-emphasis on strategies for getting a complete count in the first place. As a consequence, certain issues that turned out to be of crucial importance on the ground received only cursory treatment in training. With the benefit of hindsight, I will focus on the most significant of these: the question of how to treat ‘persons temporarily absent’ (PTA).
The intention of the ABS in 2006 was to carry out—as far as possible—a de facto count of the people in dwellings in discrete Indigenous communities. From past experience, however, it was known that some exceptions would have to be made, to cover people who were absent temporarily and who might not be counted elsewhere. In the Census Field Officer Manual for the 2006 Census, the following instructions were given with regard to absent people and vacant dwellings:
Any people who usually live in the dwelling but are away at the time of the Census should be listed in the Table for Persons Temporarily Absent on page 3 of the Interviewer Household form, unless they are at a place where they are unlikely to be put on a Census form, such as the bush or fishing.
To avoid either double-counting or missing individuals, the following rule is applied to people who were absent from the household at the time of the census:
If they were at a place where they would have been counted on a different Census form, such as a town or another community, the person should only be included in the Table for Persons Temporarily Absent.
If they were at a place where they would not have been counted, such as camping in the bush, travelling, or on other cultural or business activities, the person should be listed on page 4 with the people staying in the dwelling at the time of the Census, and all other questions in the form should be answered for the person as though they were there.
If the Collector-Interviewer comes across any dwellings in their workload where no one is staying, they should complete the front page and questions 2–3 of the Interviewer Household Form. The rest of the form is left blank.
In the CFO training notes for the training of the collector-interviewers, a further distinction is made:
Explain: If you come across a dwelling that no one lives in (an empty place), you should fill in an Interviewer Household Form for that dwelling, answering questions 2 and 3 only. This includes dwellings that are unoccupied for cultural reasons.
Explain: If you visit a dwelling where people usually live and find there is no one home, try to find out whether they will be back before the Census is finished. If they do come back, complete the forms. If no one returns during the Census time, the dwelling is counted as an unoccupied dwelling (empty place).
During their training, the CFOs flagged PTA and the rules for dealing with them as something that was worrying them. One of them said, ‘This is the issue that has been jumping off the page for me’, and they all wanted to spend more time on this topic. Their main concern was how one could know for sure that a PTA had been counted elsewhere, given the rolling nature of the count. They pointed out that the local knowledge of the CCs and CIs would be of little use here, since there was no way those people could know whether or not someone had been counted elsewhere. The general advice given was, if in doubt, include the person in the main part of the form as if they were present, and get as many details as possible for them. In the event, not a great deal of time was spent on the topic, because, in the absence of a detailed compendium of the kinds of scenarios that might occur, the discussion could not be framed in a useful way. There just seemed to be endless possible scenarios, each less clear-cut than the last.
The case-study chapters (3–6) show that the CFOs were right to be worried about this question. In the training of CCs and CIs that I observed in Arnhem Land, the CFO was not able to articulate a plan of action for the CCs and the CIs that was any clearer than the instructions he had received himself. As is argued in Appendix B, the design of the form itself tended to lead CCs and CIs to leave PTAs out of the main body of the form, and the lack of training on how to make judgments to do otherwise certainly compounded this tendency. I return to the consequences in the discussion of the post-count phase at the CMU (Chapter 7).
In talking to the CFOs about logistics, the CMU staff was constrained by the time frames that had been set by the ABS in Canberra. The CFOs were expected to have the CC and CI training, the count itself and the post-count checking completed and all the forms back to Darwin in six to seven weeks (10 July–28 August). After their CHINS experience, it was clear that most of them considered this to be unrealistic—and, as it turned out, they were right. The CMU staff was in a position in which they had to maintain the idea that everything could be done realistically within the time frame that had been set. It was clear, however, that they also had their doubts: it was said that the most important thing was to get as good a count as possible, and that if the worst came to the worst that took precedence over deadlines. They had also put into place ‘emergency backup’ provisions, with a ‘floating’ CFO who could be called in to help in areas where things were going slowly. The net effect was a tendency to curtail discussion about possible problems and to ‘focus on the positives’ to a somewhat sceptical audience of CFOs. I felt this situation is hard on the CMU staff and the CFOs, and that Canberra needs to take on board the fact that the IES—as it is presently organised—cannot be undertaken in the envisaged time frame.