A few weeks ago I updated my ongoing analysis of the RQ polling data and did some predicting in the forum. There was nothing new in what the graphs showed though so I didn’t even mention it here. I have put some more thought into what this data means in terms of what might happen to the wait and have come up with some thoughts
Ok first things first. We have this graph which is all we have for completely factual information:
All of the sites out there that you can plug your LID into and get a wait estimate are extrapolating from this graph. Using this graph for predicting future waits, particularly longer term waits is completely meaningless in my opinion.
So do we have anything better?
Well for short term predictions RQ’s poll data is about as good as we have. Using her poll data I produced this graph.
This is identical to the approach RQ has taken except I have done it graphically and arbitrarily selected different prediction categories with results as follows:
So at the current pace (red line) we get cut off dates of: Jan 27, Feb 9, Feb 17, Feb 27, Mar 5 & Mar 9
For the worse or slower pace (gold line) we get cut off dates of: Jan 25, Feb 6, Feb 13, Feb 17, Feb 24 & Mar 1
For the better or faster pace (green line) we get cut off dates of: Feb 6, Feb 16, Feb 28, Mar 9, and then we run out of poll data.
Ok so nothing new there either. What about longer term predictions. In my opinion there are just too many unknown variables to make meaningful predictions, but let me try to demonstrate why.
One of the things that I did with RQ’s poll data is I took all the daily polls she has done and using the overlapping dates in each poll adjusted all the previous polls to match the current poll. This isn’t statistically perfect (far from it), but there is enough overlap in each poll that I think the data does have some value. The result of this gives the blue part of the line in the graph above and if I show all the data you get this graph:
In theory this graph shows the trend in actual NSN referrals the CCAA has been sending out over the past two years or so. I don’t think it is completely accurate but it is in the ball park anyway. Now we finally get to my point about long term predictions. Using this graph, what would it look like if the CCAA went back to referring full months again immediately? It would look like this:
So is going to happen? Most likely not of course, but I think it is interesting to see how this looks in the context of where they have been over the last two years. It gets more interesting if you consider that this is showing what would need to happen to cover a a months worth of LID’s in what we believe are still fairly large months. Imagine we are a year down the road and they are referring May 06 instead of March 06. What if the numbers that month are 50 or 60% of March 06 numbers? With what we know from the polling and with additional attrition is that unreasonable? What if post olympics there is a 40% increase in the number of children being referred? Is that so hard to imagine? So what would the graph look like if that happened? Well if you move the red line down by a factor of 60% and the blue line up by 40% you get this:
What does this tell me? Well simply that it is not radical thinking to think that at this time next year they could be doing full months again and the wait will have stabilized at between 3 and 4 years.
Ultimately though my point is that we simply do not know what is going to happen with the wait. Sure it is possible that the wait could stabilize as I have shown above, but personally I think it is also quite possible that the program will close completely and people currently in line will never get NSN referrals ever. :dontknow:
Tags: China by Windthrow
4 Comments »