• Hello Members, This forums is for DV lottery visas only. For other immigration related questions, please go to our forums home page, find the related forum and post it there.

For those that like to argue about statistics...

Britsimon

Britsimon

Super Moderator
One of the most important unknowns is how many of the selectees will return their forms, how many will be denied and how many will get their visas. So the selectee to visa issued ratio (which varies by country and region) is important.

USCIS PUblish data about that at the bottom of this page - http://travel.state.gov/visa/immigrants/types/types_1322.html


I've looked at the three years 2011, 2012 and 2013 to see if history could tell us anything.

For 2013, we have the published selectee counts but we do not have the official visas issued numbers. We do have the CEAC data but it is missing some cases that never got in to CEAC AND it is missing the AOS cases. However, what is there is very accurate and verifiable.

2012 was a horrendous year for the lottery. The draw was made, people were notified and then they realised it was an invalid draw, so they threw away the results and redrew the winners around July 2011. That resulted in a very low response rate (I believe because many people didn't get to hear about the redraw and didn't recheck their numbers). Less than 35k visas were issued - really this year should NOT be used for analysis.

2011 is the most recent "reliable" year. From that year we see a standard number of selectees (with selectee ratios somewhat similar to this year) and just over 51k visas were issued.

I have gathered the selectees per region and the results for visas issued. I have shown 2011 data but also for comparison the 2012 data and then 2013 thown a couple of ways. Neither the 2012 nor the 2013 examples are worth relying on, in my opinion - but interesting to have them. The last column that shows the demand, shows how "oversubscribed" the lottery is this year. The max number for the global total cannot be more than 55,000 so unless 2012 happens again (which is ridiculous) there will be a shortfall. I will add my predictions for regional quotas later...

The data is shown here:-
https://docs.google.com/spreadsheet...ZnlmcU1id1Vsd3F6Qm9SQThSYUE&usp=sharing#gid=0
 
britsimon said:
One of the most important unknowns is how many of the selectees will return their forms, how many will be denied and how many will get their visas. So the selectee to visa issued ratio (which varies by country and region) is important.

USCIS PUblish data about that at the bottom of this page - http://travel.state.gov/visa/immigrants/types/types_1322.html


I've looked at the three years 2011, 2012 and 2013 to see if history could tell us anything.

For 2013, we have the published selectee counts but we do not have the official visas issued numbers. We do have the CEAC data but it is missing some cases that never got in to CEAC AND it is missing the AOS cases. However, what is there is very accurate and verifiable.

2012 was a horrendous year for the lottery. The draw was made, people were notified and then they realised it was an invalid draw, so they threw away the results and redrew the winners around July 2011. That resulted in a very low response rate (I believe because many people didn't get to hear about the redraw and didn't recheck their numbers). Less than 35k visas were issued - really this year should NOT be used for analysis.

2011 is the most recent "reliable" year. From that year we see a standard number of selectees (with selectee ratios somewhat similar to this year) and just over 51k visas were issued.

I have gathered the selectees per region and the results for visas issued. I have shown 2011 data but also for comparison the 2012 data and then 2013 thown a couple of ways. Neither the 2012 nor the 2013 examples are worth relying on, in my opinion - but interesting to have them. The last column that shows the demand, shows how "oversubscribed" the lottery is this year. The max number for the global total cannot be more than 55,000 so unless 2012 happens again (which is ridiculous) there will be a shortfall. I will add my predictions for regional quotas later...

The data is shown here:-
https://docs.google.com/spreadsheet...ZnlmcU1id1Vsd3F6Qm9SQThSYUE&usp=sharing#gid=0
Click to expand...
Basically, your analysis for DV-2012 confirms what I call "Sloner axiom". That is what he pointed out long time ago - the large amount of selectees in DV-2014 had the purpose of "insurance" for events/results of DV-2012. Even if that happens again in DV-2014, there is no quota underfilling.
 
Simon, I can see that you have done your analysis at the region level. This overlooks the "shifting mix" within the region from one year to another.

Let's assume the success rate of each country remains constant from 2011 to 2014. However, you may now have more selectees from a country with a high success rate. In this case the overall success rate will increase, because the weights of those countries change; even if each individual country's success rate remains as before.

What you should be doing is to apply the 2011_country_success_ratio to 2014_country_selectee_count, and then aggregate. This will automatically take care of the "shifting mix".

I hope it makes sense.
 
c1984 said:
Simon, I can see that you have done your analysis at the region level. This overlooks the "shifting mix" within the region from one year to another.

Let's assume the success rate of each country remains constant from 2011 to 2014. However, you may now have more selectees from a country with a high success rate. In this case the overall success rate will increase, because the weights of those countries change; even if each individual country's success rate remains as before.

What you should be doing is to apply the 2011_country_success_ratio to 2014_country_selectee_count, and then aggregate. This will automatically take care of the "shifting mix".

I hope it makes sense.
Click to expand...

Yes agreed. I had thought of that - and the data is available at country level BUT it was more work and I thought it was going to result in relatively small differences. I might try it on a region and see what happens.

There are other variables that factor into this. One of them is how keen people are to return their forms (i.e. how appealing is living in the USA). Post 9/11 and for a few years after (while the USA pursued wars that were increasingly unpopular) there was a shift in sentiment against the USA. Britain suffered in the same way also (anyone who watched Eurosvision song contest could attest to that!). So - the take up in the years from 2004 to 2009 or so was probably affected by that. 2008 had the economic disasters and again that changed sentiment. The improving economy in the States is changing things once again. ALL of that affects the ratio. More people will return their forms in 2014 than in 2011. That is evidenced by the H1 allocation over the last few years. Anyway, because of that, the numbers are only a guess and so I decided against worrying about the shifting mix.
 
raevsky said:
Basically, your analysis for DV-2012 confirms what I call "Sloner axiom". That is what he pointed out long time ago - the large amount of selectees in DV-2014 had the purpose of "insurance" for events/results of DV-2012. Even if that happens again in DV-2014, there is no quota underfilling.
Click to expand...

Yep understood. 2012 won't happen again so I agree that might have been a driver for the over-reaction but sadly it will just mean more disappointed people...
 
It seems that in the past the DOS tried to pick the number of selectees in a way that really minimizes selectees left without visas. It is very compassionate of them, as getting selected & approved but not getting a visa is a horrible thing. However, if they select too few then they're wasting the quota and not doing immigrants a favor either. Now they seem to shift the policy to select a good bit too many, which means there's virtually no danger of unused quota but guarantees that many selectees will be left out. I can't blame them, as they say it everywhere that getting selected doesn't guarantee you a visa.
 
In another thread, Sloner estimated the split based on “Slonermetrics”. He said:

“Consider: 20 + 20 + 10 + 2.8 = 52800. I do not see the excess of the quota.”

It made me want to look at the splits and whether the split is influenced by the selectee split or not.

Based on the 2013 selectee numbers (which show round number results - clearly someone trying to determine a split) I believe the quota is administered on gross selectee counts. In 2013 it was 50:15:31:2:2 (AF:AS:EU:OC:SA). If USCIS were ignoring the different success rates among regions (as I believe they do) that would have ended up approximately 25:7.5:15.5:1:1 which is not far off the split that has actually happened (according to CEAC data and estimated for missing data). Because of success rates, AF has done worse than that and AS has performed better - the other three regions are pretty close.

If you do the same exercise with 2014 the selectee split is 62:23:47:4.2:4.6. Taking the whole 55,000 (i.e. no NACARA) and NOT allowing for higher success rate that would result in a split of 24200:9100:18200:1600:1800. I think, as before AF and AS will experience different success rates, so I would adjust that to a projection for actual split of 23000:10300:18200:1600:1800 - these being based on the MAXIMUM quota with no allowance for NACARA.
 
vladek15425 said:
Applying 2011 succes rate for 2014 is really a brilliant idea, because numbers
Don't lie.
Click to expand...

I can't take all the credit. It is the Sloner approved method - BUT he took 2012 results - which shows why he is so positive about all getting interviews. Trouble is 2012 is the worst year to use. Ever.
 
Applying 2012 succes rate on any othe dv lottery will give you visa for everyone :)
That's why its sloner favorite lol
 
Trouble is 2012 is the worst year to use. Ever.
Click to expand...
Why? You have not resulted facts.
That resulted in a very low response rate (I believe because many people didn't get to hear about the redraw and didn't recheck their numbers).
Click to expand...
This is nonsense. Show me the facts.This is speculation. Who said that? Error was 1 May 2011. July 15, 2011, everything was fine.
 
Last edited by a moderator:
Sloner said:
Why? You have not resulted facts.

This is nonsense. Show me the facts.This is speculation. Who said that? Error was 1 May 2011. July 15, 2011, everything was fine.
Click to expand...

Visa bulletin for May 2012 has special section E that was never a part of visa bulletin before or after (so far).
http://travel.state.gov/visa/bulletin/bulletin_5692.html
That is how they admitted the problem first. That is very contrary to what DOS lawyer stated earlier, in March 2012 brief for the court (unusually high number of visa applications received in KCC)
Chronologically that is the first proof of the failure DOS had in DV-2012. The second one is the visa statistics for DV-2012 stating the number of visas issued.
 
britsimon said:
Yes agreed. I had thought of that - and the data is available at country level BUT it was more work and I thought it was going to result in relatively small differences. I might try it on a region and see what happens.

There are other variables that factor into this. One of them is how keen people are to return their forms (i.e. how appealing is living in the USA). Post 9/11 and for a few years after (while the USA pursued wars that were increasingly unpopular) there was a shift in sentiment against the USA. Britain suffered in the same way also (anyone who watched Eurosvision song contest could attest to that!). So - the take up in the years from 2004 to 2009 or so was probably affected by that. 2008 had the economic disasters and again that changed sentiment. The improving economy in the States is changing things once again. ALL of that affects the ratio. More people will return their forms in 2014 than in 2011. That is evidenced by the H1 allocation over the last few years. Anyway, because of that, the numbers are only a guess and so I decided against worrying about the shifting mix.
Click to expand...

Agreed. With what I'm suggesting we would still hold the assumption that the return/success rate will remain constant (you had assumed it at region level, I'm suggesting taking it at country level), which is the best that we can do keeping in mind all the drivers you had listed.

However, the country mix of selectees is something that we clearly see is changing from one year to another and we can conveniently account for (see the change in the number of selectees for Nepal and Iran from 2011 to 2014 for example).
 
Last edited by a moderator:
Sloner said:
Why? You have not resulted facts.

This is nonsense. Show me the facts.This is speculation. Who said that? Error was 1 May 2011. July 15, 2011, everything was fine.
Click to expand...

We keep coming back to this.

Some people checked their status on May 1st and saw that they were not selected. Many just said "oh well", moved on , forgot about the whole thing, were not aware of a re-draw and never checked their status on July 1st because they thought they were done.

Many that were selected on July 1st did not learn of their win. These people under normal circumstances could potentially pursue their case and send forms to KCC, but they didn't. So the response rate for 2012 was significantly lower than the other years. That's why 2012 is not "normal".

You may see it as speculation, but it is a lot closer to reality that many other speculations your have come to accept as fact.
 
We keep coming back to this.

Some people checked their status on May 1st and saw that they were not selected. Many just said "oh well", moved on , forgot about the whole thing, were not aware of a re-draw and never checked their status on July 1st because they thought they were done.

Many that were selected on July 1st did not learn of their win. These people under normal circumstances could potentially pursue their case and send forms to KCC, but they didn't. So the response rate for 2012 was significantly lower than the other years. That's why 2012 is not "normal".

You may see it as speculation, but it is a lot closer to reality that many other speculations your have come to accept as fact.
Click to expand...
This is nonsense. Conversely. Auditors should be more. How could they forget? Second chance to win.
 
They did not forget. They did not know about the replacement draw. Not a lot of people follow this topic on a daily basis.
 
Guys I'm not sure what we can deduce out of these numbers. If the visas issued include derivatives then comparing with sample of selected winners make less sense as the two populations ie. selectees and visa holders will be different. Also we don't know the maturation rate of the selectee population and we'd should assume that follow through of selectees decreases over time.
 
You must log in or register to reply here.
Top