The Australian has a story saying that Labor is in trouble in Queensland. This is based on an aggregation of recent Newspolls into 3 month chunks, giving Newspoll decent sample sizes in each state (except Tasmania). Some quick reactions.

At the last election, in 2010, Labor got 33.6% of 1st preferences cast in Qld. The Coalition got 47.4%. The aggregation of April to June Newspolls has the corresponding numbers at 31% and 48%, respectively. That is not a big change, and certainly not “statistically significant” on the Coalition side of the ledger, although probably getting close on the Labor side. The bigger story might be the way the numbers for “Others” are bouncing around: 8.1% at the 2010 election, and then 13%, 14% and 11% in the Newspoll aggregations. Where those votes and their preferences wind up going will be quite consequential.

Interestingly, Labor was at 31% in the Feb-March 2011 Newspolls too. So the “Labor on the nose in QLD” story is hardly new, nor especially striking. I agree that a seat like Lilley (Wayne Swan’s seat, and taking in my old northside of Brisbane stomping ground) hasn’t got a lot of slack in it for Labor (and indeed, has changed hands when Labor is really on the nose). But the QLD numbers aren’t exactly jumping off the page.

Bigger differences from 2010 lie elsewhere: NSW 37.7 (2010 actual) to 30 (Apr-Jun 2011 Newspolls) for a 7.7 percentage point fall. VIC, 42.8 to 37, a 5.8 fall. SA: 40.7 to 34, a 6.7 fall. WA: 31.2 to 27, a 4.2 fall. Qld: 33.6 to 31, 2.6, which is the smallest of the falls in the state-by-state break-outs.

Put differently, look at the graphical depiction of the data I’ve made below; after seeing that, how are the QLD numbers the story? If the NSW numbers are even close to being right, then you can forget QLD, it won’t matter.

Now that classes are over, I took a little time to update my scripts that update the analysis of Congressional roll calls in close to real time.   Links appear at the top of the blog.   As of about 15 minutes ago, we’re up to 77 non-unanimous roll calls in the 112th Senate.   The House has 474 non-unanimous roll calls under its belt.

I’m presenting estimates of legislators’ “ideal points” and 95% credible intervals (from a model that fits just a single underlying dimension to the roll calls) both graphically (House/Senate) and in CSV.  I also present scatterplots (and loess smoothing) of the estimated ideal points against a crude (but useful) measure of preferences in the legislators’ district/state, Obama vote share in the 2008 election (House/Senate). I’ve also got a SVG with rollovers for the dense House scatterplot, using the RSVGTipsDevice package, but the resulting SVG breaks in Chrome.

I’m scraping the roll calls and some meta data from the House and Senate sites, using the parsing in R’s XML package (which I’m finally understanding how to use effectively).   Analysis of the roll calls is via the ideal function in (my) R package, pscl.

Quite aside from the methodology/technology, the substantive story is very much business as usual: zero partisan overlap in the recovered ideal point estimates. About 1 to 1.5 standard deviations of the ideal point distribution separate the ideal points of Democrats and Republicans among districts/states that split 50-50 Obama/McCain in 2008.

The other striking feature of the data is how few Democrats remain in the 112th House in districts where McCain beat Obama: I count 12 such seats.

From Sam Gubins, President and Editor-in-Chief of Annual Reviews, Inc, publisher of the Annual Review of Political Science:

Dear Editors of the Annual Review of Political Science:

From strength to strength. The Thomson Reuters Journal Citation Reports (JCR), the annual report card of impact factors, was released today.

The Annual Review of Political Science is the #2 impact factor journal among 179 journals in the category POLITCIAL SCIENCE. It has an impact factor of 3.0, just behind APSR with 3.278.

There are 2,678 in the JCR, Social Science Edition. The Annual Review of Political Science enjoys an impact factor rank of #154 aong these journals, placing it in the top 6% of all journals in the social sciences.

This is a truly extraordinary success. You should be very proud of what you have accomplished.

Congratulations.

Sam

President & Editor-in-Chief
Annual Reviews
4139 El Camino Way
Palo Alto, CA 94306

http://www.AnnualReviews.org

SCOTUS Cj Roberts: “Laws like Arizona’s matching funds provision that inhibit robust and wide-open political debate without sufficient justification cannot stand…”

Its probably unfair to pick on this particular line of Roberts’ opinion, but just how does “topping up” a candidate’s spending with public funds “inhibit robust and wide-open political debate…”? This — I gather — was the point of Kagan’s dissent.

Same 5-4 split as Citizens United, similar logic, same substantive result…

I snuck away from a conference in Dublin for a few hours to grab these photos of my ancestors’ tombstones in Dunmore East, Co. Kilkenny. I found the tombstone of my great-great-great grandfather, Patrick Jackman (ca. 1782 – 1847).

I found an even older Jackman tombstone, for a Michael Jackman (1763-1820), who might have been my 4th great grandfather (we don’t know).

My thanks to Kathleen Frailey Puls who clued me into the fact that the Dunmore East cemetery contained these Jackman tombstones.

View Larger Map

The Nielsen poll with Labor on 27% 1st preference vote share has caused quite a stir. The poll I ran for the USSC in mid-May had Labor on 30%, which made me a little worried about bias in the sample (although usually bias tends to run a little pro-Labor with on-line polling etc).

Turns out we got numbers very close to what Nielsen got a couple of weeks later, using a different methodology etc.
Nice…

Stanford has a great Stats Dept: this is what passes for controversy there. I cut and paste a redacted version of some traffic on the Stats Dept email list…
————————

I believe this problem has arisen before. (See Goldilocks and the three bears.)
Hope this works out as well.

On 6/15/11 6:35 PM, [name deleted] wrote:
Dear Stats,

I left a plate of pasta from today’s lunch in the refrigerator that I
was saving for dinner with my name clearly marked on it, and now it’s
gone. It’s not a big deal, but I hope that this is not the beginning
of a trend of people taking other people’s food….

Also, I left a white bowl (with a blue rim) in the lounge about a week
ago and it’s no longer there. I assume that someone has been using it
in the meantime, which is great, but I would really like it back if
you’re done with it. Thanks!

Sweave source for the poll report for those who expressed some interest.

You’ll also need this file of R function definitions, utilities.R.

I also wrote a little shell script that calls Sweave and xelatex etc, hacking the Sweave.sh script that ships with R.

I did quite a lot of Australian media last week with the 9/11 Summit and the release of our poll.

Some video highlights: ABC The Drum; 6.30 with George Negus; ABC News Breakfast.

We also got an op-ed up in the Financial Review (behind a pay wall), plus a lot of other media. A big thank you to everyone at the United States Studies Centre, but especially Nina, Annabel and Suze on the media side of things.

To the Moon and back, twice… I get some upgrade awards and lifetime Premier Exec status.