An interesting report from the (Australian) ABC:link here. Interesting for linguistic and well as social reasons.
Note headline:
ABS confident all NT covered by census.
Now note first line:
The Australian Bureau of Statistics (ABS) says it is confident most remote Indigenous communities in the Territory will be counted in this year’s census.
Now, in most dialects of English, all and most do not have the same meaning… My classificatory brother Wamut posted that Ngukurr didn’t get their census forms, amongst other communities. Any news, wawa?
Anggarrgoon 
yapa,
the latest info I have is that Ngukurr received forms but they’re all in boxes at the council office waiting for someone from ABS to tell them what to do with them/help them. So I guess someone will come along at some stage.
Funny though, I don’t know how accurate it can be. Ngukurr average population is a bit over 1000. But if there’s a funeral on, well suddenly it’ll be more like 1500. Or if there’s a funeral in a neighbouring community, our population will seem rather small. People move around so much I don’t know how ABS can think they can catch everybody.
Well, if they catch a few of us at Ngukurr it’ll be a start.
Socially, what a shame! Linguistically, maybe they omitted “the” before “most” so that it would be referring to the remoteness, not the percentage? They repeat “all” later in the article, so it didn’t seem like they were backing away from the claim made in the title.
If they did the census on a single night they’d catch everyone somewhere – and isn’t there a question about whether you’re in your usual place of residence? Not that that’s a particularly useful question in some cases..
There is an issue in the US about correcting for undercounting which is a political issue – on the assumption that filling out census forms is positively correlated with wealth, and so is voting Republican, then applying statistical corrections for this in electoral districting means more safe Democratic districts. I realise this doesn’t apply federally in Australia given the limited (one? two?) number of seats in the House for the NT (or is this an issue in where the boundaries of seats like Kalgoorlie in WA or whatever seat covers Cape York in Qld are drawn, ie how many mining towns and how many indigenous communities there are?), but does the census affect allocation of anything else? CDEP money? Education resources? Telecommunication being provided to communities with certain minimum populations? If so, and if communities are being undercounted, it seems like there might be a prima facie case for legal action against the ABS under human rights and equal opportunity legislation if an appropriate effort isn’t being made to count everyone. Or if not against the ABS, against agencies that don’t adequately correct for undercounting in allocating resources.
I know someone who flew back to Australia from Europe the day of the census, and therefore had to answer that their commute to work began in Helsinki.
Atsic funding was partially based on census figures within individual regions. Not sure what’s happening with that now. Education resources, afaik, are based on enrollment counts, but they have their own problems. CDEP positions are also based on community size. It also has implications for data availability – the ABS won’t release statistics for areas with less than a certain number of responses (can’t remember how much) because of confidentiality concerns – when a community like One Arm Point misses out, that also causes all the outstations data to be blocked from release too, or combined with other areas.
Note first that in most media organizations the person responsible for the story is not the same as the person responsible for the headline. The former is usually called a writer, the latter a sub-editor. So, it is probably not the fault of the writer that the headline here disagrees with the story. It is almost certainly not the fault of the ABS, unless their press release had this headline or these words in it.
Secondly, it is no simple matter to make certain you have counted everyone in a national census. It is easy say that if it is all done on one night, then everyone must be somewhere. But what of people traveling on that night? What of the homeless? What of people camping in a national park, sleeping on a beach, parked by the side of the road in a remote area, etc? How do you ensure that all these people receive, complete and return their census forms? Not an easy matter, and one which national statisticians usually devote considerable effort, time and resources to dealing with.
My griping about the headline was directed at the ABC, not the ABS.
We’re not implying it’sa simple matter, but there’s a real legal and moral issue here. Based on the anecdotal reports, Aboriginal communities are frequently missed in census counting. That’s illegal. It has potential distorting effects elsewhere.