Poll Numbers – Poll Closed Feb 2012
I put the poll numbers up on the Poll Results page yesterday, but I’m putting them out here as well. The reversed data points represent weekends. For 2009 and beyond I’ve just listed it as LID’s per month, those are the grey data points.
As a reminder, I won’t use the new poll right away. I’ll gradually begin integrating it into my formulas, but for the first few months it will have ten percent or less of the weighted averages I use.
After spending some time playing around with days referred per month versus LIDs referred per month, I still believe the LID polls are our most important forecasting tool. I’ll have more on that when I put the Projections post together.
Edited to add:
I don’t use any one poll for my projections. I use a weighted average formula that includes all relevant polls. Here is a side by side comparison that may put things a bit more into perspective.
Remember, people are more likely to record their LID in a poll shortly after LID, and then again as their referral date draws closer. There is a period of time in between where many people need to step away, and we don’t see as high of a percentage voting in polls during that time. Our overall percentage is around thirty percent, but for people very close to referral we are a good bit higher. Subtract attrition, add back in because these people are seeing the light at the end of the tunnel, and it looks about right to me.
.
February 23rd, 2012 at 8:58 am
It appears that large numbers of September LIDers disappeared between the prior poll and this poll. Any idea as to what happened? I don’t think it would all be attributable to double counting.
February 23rd, 2012 at 9:43 am
Looking at the last five polls side by side, I’d say it looks fairly consistent across the board. I mean, sure, we’ve had a lot of attrition, but the weighted averages I use still show the same big days and small days. Since we’re only getting around a third of respondents, and since some days we may be getting more than a third and other days less than a third… Also, keep in mind that some agencies have been a lot more proactive about switching families to the SN program, so the days with a drastic drop are probably days that one (or maybe several) of the proactive agencies had a lot of people logged in.
Give me a bit and I’ll do a screenshot and post it.
February 23rd, 2012 at 4:16 pm
I honestly think they no longer look at numbers of files logged-in. It is simply a calculation of days, which is why it just so happens that they seem to do about 2 months of LIDs each year, excluding March of 2006, of course. This can no be a coincidence! Therefore, we can expect a maximum average of 5 days per month over the course of a year. That seems to be the baseline and then things can only go down from there, because it would be an absolute miracle if they ever referred any faster than that. I look forward to the projections post to hear why you think the number of LIDs is still a valid predictor.
February 23rd, 2012 at 4:18 pm
i think i missed the 2011 US adoption statistics? sorry if it was already discussed.
the site also has a graph now and they provide ages of children now which i don’t recall seeing before.
http://adoption.state.gov/about_us/statistics.php
so 2011 down 26% vs. 2010, but there were more adoptions in 2010 vs. 2009….??
total adoptions to US:
2011 – 2,587
2010 – 3,401
2009 – 3,000
2008 – 3,912
2007 – 5,453
2006 – 6,492
2005 – 7,903
2004 – 7,038
2003 – 6,857
2002 – 6,116
2001 – 4,705
2000 – 5,058
1999 – 4,108
February 23rd, 2012 at 10:11 pm
There is a thread on the 2011 stats I started in the forum a while back. The 2010 vs 2009 stat is misleading, because the NSN adoptions were still dropping year over year. The difference was due to SN adoptions.
February 24th, 2012 at 8:33 pm
so the black squares are marked because they are a weekend, and that means that it probably isn’t a correct Log in date? do you have a guess as to the agency having it wrong? Will the CCAA place the family forward or backward in the wait, one day can make a big difference these days?
February 24th, 2012 at 8:44 pm
They work on weekends sometimes – especially before or after a holiday. Also, there are no sick days, if someone is out because they are sick they have to come in on a weekend and make the day up.
I use all kinds of things to try to see patterns in the numbers, that’s just one of them.