jackman.stanford.edu/blog
bannerImage

arb this (again)

Thursday May 31, 2012

Filed under: Australian Politics — jackman @ 5:26 pm

The Coalition is at 1.17 at Centrebet; the ALP is at 7.00 at sportsbet.

1/1.17 + 1/7 = .9976 < 1. But not by much.

Do ya best.

Comments (1)

Australians do it longer…

Sunday May 27, 2012

Filed under: flight nerdery — jackman @ 3:30 pm

Janet got back this morning from visiting family in Brisbane. She went on Star Alliance miles, with the return we could find being via Singapore and Narita (not the most direct BNE-SFO routing).

Janet flew over 20,000 miles flown on this trip. That is the equivalent of 4 SFO-JFK roundtrips, or about 2 SFO-LHR roundtrips. And this is the West coast of the US to the East coast of Australia. NYC – Perth via DFW and BNE is 11,940 miles, which is an awful long way for a family visit.

Interesting coincidence that BNE is basically on the GC route between AKL and SIN, and that NRT is close to the SIN-SFO GC.

Comments Off

political/economic/demographic indicators, by state, by year

Friday May 25, 2012

Filed under: politics,statistics — jackman @ 10:28 am

“Good luck with that…”

How many times do you see a post to a list like PolMeth where someone is looking for data on x, by state, by year? What would see if you looked at the helpful follow up posts? This x is over there, but only goes back annually to 2000, then in five year intervals to 1980. Some other variable lives somewhere else, but only for Census years, but does goes back to 1900. Or, the Census has x in a flat file ready for download, but only for 2010. Really quite a mess… Lots of scanning or key-punching, scraping and merging. And probably much wasted effort, when one thinks about people having to do this for themselves, across the profession..

Comparative politics has resources like the Penn World Table, or the Luxembourg Income Survey, etc. IR has COW, MAR, Polity, Freedom House. The World Bank. The IMF. The OECD. The UN.

Why is data about the American states so much harder to get? Or am I missing something?

And please no quips about this all being yet another manifestation of the genius of American federalism…

Comments (1)

Palo Alto executive arrested for Lego theft

Tuesday May 22, 2012

Filed under: general — jackman @ 9:42 pm

“And now for something completely different…”

Click on the thumbnail for higher-res JPG.

Comments (1)

under-coverage bias and Kish on field enumeration for area-based samples

Saturday May 19, 2012

Filed under: ANES,statistics — jackman @ 5:51 pm

As one of the PIs of the 2012 ANES, I gained some exposure to the nitty-gritty of how area probability samples work in practice. We’re using an ABS-frame (the USPS Delivery Sequence File), which we will supplement with some field enumeration in Census tracts where the DSF is thought to be subject to a reasonable amount of under-coverage.

What I’ve learned thus far:

(1) Kish’s Survey Sampling remains something of a bible for practitioners.

(2) This book really is for practitioners, with large sections devoted to what actually occurs in the field, how to walk around a block, listing addresses, etc. Its odd to read this stuff. I mean the rubber does have to hit the road at some point. But so much of it seems a little, well, folksy and even ad hoc, unless I’m missing other parts of the book where the underlying rationales are more rigorously explicated. I guess it has to be that way, when you are trying to keep things simple for the non-statistician field workers.

(3) Take this, the case of how to augment a listing of dwellings when the field worker encounters dwellings not on the list (in our case this would be finding dwellings not on the DSF adjacent to a dwelling sampled from the DSF). From pp341-2 of Survey Sampling (JPGs below are clickable thumbnails), something of a “how-to” guide for the Half Open Interval procedure:

Take all unlisted dwellings if there are less than 5 of them? What is special about 5? If 5 or more, “write the office quickly” (presumably today, you’d call) and “wait for instructions”. And what, exactly, will those instructions be?

I’m sure there is some well-worked out basis for these recommendations somewhere, perhaps elsewhere in the book. At p56 Kish says that the “missed [but discovered] elements receive the same probability of selection as the pre-specified unique listings”.

Ok, but might you get too many unlisted dwelling this way? Interviewer workload becomes an issue then, which where I guess “no more than 5″ might come from.

But could you exploit whatever prior information about DSF under-coverage specific to the locality you’re working in? And at that point I guess you might be stratifying dwellings in a given geographic unit into listed and unlisted and heading towards a dual frame design etc.

Sub-sampling seems another idea: e.g., the design calls for r attempted interviews in a given locale. We sample r listed dwellings in the locale from, say, the DSF; field enumeration adds k to the frame around the listed r, we attempt interviews at r SWOR from the r+k? This keeps the IWR workload down to r attempted interviews and the selection probabilities are “known”.

The literature on snowball or “respondent-driven” sampling in social network land must have some relevant ideas here too. Hitting r listed dwellings and then looking around for unlisted dwellings seems a lot like what goes on with sampling on networks for “hidden” populations etc.

Finally – I have to note that this stuff really is probably 2nd order at best. We’re doing our best on the design for the in-person components of ANES 2012, I think. But there is this big scary monster out there, waiting for us in the Fall when we go into the field, and its name is non-response. As a source of bias this has to be 10x what we’re looking at from DSF under-coverage.

Comments Off

Thought of the day re Political Science at NSF

Friday May 11, 2012

Filed under: politics — jackman @ 5:05 pm

The fight to save the American Community Survey crowds out the voices trying to keep NSF funding for political science…

Comments Off

House votes to cut NSF political science funding

Wednesday May 9, 2012

Filed under: politics — jackman @ 11:45 pm

At 11.57pm tonight the House passed an amendment to HR 5236 cutting NSF funding to political science, 218-208. 5 Dems voted Aye. 27 Reps voted Nay.

Of 3 amendments moved today by Jeff Flake on HR 5326, this one got up. We just learned a lot about preferences for $9M/yr of political science funding vs cutting billions from NSF funding in the aggregate.

At this point we’re hoping the Senate and conference turns this around. Write your Senators…

See my earlier post today on this.

Oh: and they freaking well wiped out the American Community Survey too with this rollcall. Amendment of Daniel Webster (Rep FL-8).

Comments (1)

another threat to NSF…

Filed under: politics,statistics — jackman @ 10:50 am

Email arrived overnight with news that the National Science Foundation’s support of political science research might be under threat. Specifically,

APSA [the American Political Science Association] has learned that Representative Jeff Flake (AZ) may imminently introduce an amendment to the NSF appropriations bill now on the House floor (HR 5326: the Commerce, Justice, Science, and Related Agencies Appropriations Act, 2013) to defund the political science program at the NSF.

Flake is running for the AZ Senate seat opening up with Kyl’s retirement. The Republican primary in AZ is in September. Flake is drawing some opposition on his right flank, it would seem. Draw your own conclusions, perhaps sprinkle in the fact that Republican Senate aspirants find themselves in a “post-Lugar” environment, etc.

For what it is worth, Flake has a slightly more conservative voting history than we’d expect, even given that he’s in a seat that went for McCain over Obama by better than 60-40. Take out the McCain home-state effect and the “normal vote” for AZ-6 slides a little more Democratic, meaning that Flake is even a little conservative again, given the district’s presidential vote split. When you look at Flake in relation to other AZ House members (solid dots on the graph above), he’s roughly middle of the pack. This leads a little bit of credence to the “conservative, but conservative enough for the statewide Republican primary constituency?” hypothesis, and, in turn, why we’ve got some position-taking like what’re seeing.

Yesterday Flake got a vote up on the floor of the House, an amendment that he described as taking the NSF budget back to pre-stimulus levels (Congressional Record). It failed, with all Dems voting against it, but with Republicans splitting 121-112 in favor.

The roll call split Republicans pretty cleanly, with legislators’ ideal points a reasonable but not great predictor of Republican votes (area under the ROC curve is 0.945 for everyone, down to 0.858 for Republicans). The next graph shows the item-characteristic curve for the roll call, with ideal points and actual Yeas and Nays superimposed. There are plenty of mis-classifications: 22 Ayes predicted to vote Nay (prob < .5) and 25 Nays predicted to vote Yea (prob > .5), for a total of 20% mis-classified. So there is a bit more going on here than “ideology” among Republicans.

Note also that Coburn moved to kill political science funding at NSF back in 2009. I blogged on that at the time. See here (n.b., 5 Senate Dems supported Coburn’s amendment).

Comments (1)

confirming record territory in the Australian political betting markets

Monday May 7, 2012

Filed under: Australian Politics — jackman @ 11:34 am

An update, confirming we are in record territory on the betting markets (afaik). Two previous highs/lows in the national betting markets:

Centrebet had the Coalition on 5.15 on December 2, 2009, to Labor’s 1.14, for an implied probability of a Coalition win of .18. This was the day after Abbott became Leader of the FPLP. Rudd had about 200 days left as PM at that point.

Centrebet had the Coalition at 4.95 to Labor’s 1.18 on Nov 20, 2007 (Tuesday before the 2007 election), for an implied probability of a Coalition win of .19. Newspoll releaed a 54-46 2PP poll that morning, from memory.

Comments (2)

Labor out to $6.00 on Sportsbet

Sunday May 6, 2012

Filed under: Australian Politics — jackman @ 2:43 pm

This could well be the longest odds I’ve ever seen in Australian national-level political betting markets. 6.00 to 1.13 at Sportsbet at 7am this morning, Sydney time. I’ll ransack the historical data to verify that.

Others will likely follow as that 15% ROI offered by Centrebet (inter alia) for a Coalition win gets stomped on.

Comments Off
Next Page »

Powered by WordPress

Bad Behavior has blocked 3083 access attempts in the last 7 days.