NSC Processing From Rupnet

P

paulclarke1

Registered Users (C)
At the risk of being 'on topic', here is my 11/3/03 take on where the NSC is with processing cases based on Rupnet data.

From what I can see NSC is about 2/3 the way through their initial pass through Jan 02 cases. Based on the pace they worked previous months cases they should be done in another 4 -6 weeks.

This does not mean that all Jan or prior cases will be approved, just that they will have finished there first look, approved or sent RFE's to the obvious cases, and placed the rest in the too hard basket to wait until God Knows When. This first look analysis does provide any insight into how long a particular case will take after an RFE is responded to.

The current pace of activity at NSC does not seem to provide any backlog relief as they seem to have been stuck on Jan cases since the week of 5/26 (10 weeks ago). In my estimates Jan 02 first look processing will probably take 14-16 weeks in total. This means you can add another 10-12 weeks into the waiting time for cases filed since May 03.

There are some early suggestions of nascent work on Feb 02 cases, but usually I like to wait until the cases worked (Approvals + RFE's - Approved RFE's)/Total is greater than 10% to be more confident.

At this stage I do not see any evidence of the post October bounty of new NSC activitiy that we all hope for. We have seen an improving trend in the number of cases approved since August, but these seem closely linked to approval of RFE cases. The number of new cases worked seems to be remarkably stable and the approval trend would seem to suggest more of house cleaning than breaking new ground.

As a proactive statement regarding the fact that we have seen greenland script activity on cases as recent as May 02. I don't doubt that these results are real, I just doubt that these are EB cases. Otherwise why have we never seen any ownership of these fast approval cases either here or in Rupnet.
 
here is a reason,

the rupnet/immigrationportal-posters sample is too small (and possibly skewed on home country of applicant data) to be representative of the total EB population.

As for non-EB pollution of the greenland combined data, it becomes a question of figuring out what other types of I-485's there are, and how long is their processing. My understanding (which is quite limited) is that the other stuff is significantly quicker than the EB stuff (or significantly slower, ie certain FB types). I would think that if you plotted the bell curves for various types of AOS approvals, you wouldn't get much of an overlap, (ie. very unlikely for it to be in the first +/- sigma range). IMHO, looking at the funkyjunky data, the EB bellcurve is obvious in the Oct 2001 to May 2002 range, with a peak around january 2002. This would support a number of facts, like the Aug 2001 BCIS cutoff (perhaps thats the six sigma point), the high jan 2001 activity apparent on rupnet (I look at montly data, and look for 'color'--yellows and greens-- to see where the crest of the wave appears to be), and the dropoff towards May 2002 cases.

It'd be even better if we extended the funkyjunky spectrum from Oct 2001 to the present, perhaps we could even see the "Derivative AOS" bell curve in the jan 2003-jun2003 range, assuming a 7-8 month typical processing time.
 
Re: here is a reason,

Originally posted by SirZ
[B IMHO, looking at the funkyjunky data, the EB bellcurve is obvious in the Oct 2001 to May 2002 range, with a peak around january 2002. This would support a number of facts, like the Aug 2001 BCIS cutoff (perhaps thats the six sigma point), the high jan 2001 activity apparent on rupnet (I look at montly data, and look for 'color'--yellows and greens-- to see where the crest of the wave appears to be), and the dropoff towards May 2002 cases.[/B]
Click to expand...

To be able to construct a bell curve on the current processing activity, funkyjunky will have to make available his table in its delta form (say, month over month) and not as cumulative activity, which is how it is currently presented. This would be an excellent enhancement to the greenland report if funkyjunky can incorporate it; not too difficult either.

P.S.: I suppose you mean the three sigma point, as in three standard deviations on the left side from the mean.
 
sigma

You are right, I suppose its 3 sigma since I'm only referring to one side of the curve :). My company is big on Six Sigma, and we often mis-use the term to refer to the boundary, even when focusing on only one side.
 
I see your points. I would have to create a benchmark version where all the LINS are updated as of a particular date. This will be a bit of a challenge (1000 limit), but I will give it some thought and see if I can come up with an acceptable solution.

Thanks for the input
 
You must log in or register to reply here.
Top