Slashdot Mirror

Publications comprised 20% (2012) and 27% (2011) of ASA's budget, according to this audited report on page 10 of their membership magazine's June 2013 issue (pg 10) (PDF). They make a big chunk off publications, but I wouldn't say that's "most of" the revenue; membership dues accounted for 29% (2012) and 25% (2011) of revenue.

You mean like the overall long-term increase in Antarctic ice mass, despite breakups in the Western sheet?

False. Antarctic land ice mass is decreasing, and reliable estimates of Antarctic sea ice volume (or mass) aren't available.

Even if you meant to refer to Antarctic sea ice extent (not mass), you already ignored me when I told you that this is consistent with Manabe et al. 1991 page 811: " sea surface temperature hardly changes and sea ice slightly increases near the Antarctic Continent in response to the increase of atmospheric carbon dioxide."

But maybe you'll listen to the National Academy of Sciences, if you honestly don't think the National Academy of Sciences is "alarmist". Again, their recent report is educational. They address Antarctic sea ice in question 12.

The gradual, long-term non-warming that has occurred over the last 15-17 years, depending on who you ask?

Jane and Lonny Eacus have repeatedly ignored me whenever I've told you that there's been no statistically significant change in the surface warming rate. But if you honestly doesn't think the NAS is alarmist, you might learn something from their answers to questions 9 and 10. This point is particularly relevant: "More than 90% of the heat added to Earth is absorbed by the oceans and penetrates only slowly into deep water. A faster rate of heat penetration into the deeper ocean will slow the warming seen at the surface and in the atmosphere, but by itself will not change the long-term warming that will occur from a given amount of CO2."

I agree: science is a wonderful thing. You can appear to "prove" almost anything you want if you restrict your study to relatively isolated phenomena, and ignore the bigger picture.

No, that's not science the way it's practiced by the National Academy of Sciences, the National Center for Atmospheric Research, the American Geophysical Union, the American Institute of Physics, the American Physical Society, the American Meteorological Society, the American Statistical Association, the American Association for the Advancement of Science, the Federation of American Scientists, the American Quaternary Association, the American Society of Agronomy, the

Statistics. That unforgiving bitch no one asked to be invited that tries to manipulate all of our lives.

Statistics don't manipulate people - people manipulate people.

I'm the ASA. And I vote.

The Census site has a little info about this: http://www.census.gov/privacy/data_protection/statistical_safeguards.html

But more relevant is this link to the American Statistical Association, which goes into significant depth on the techniques used to protect confidentiality: http://www.amstat.org/committees/pc/index.html

On this page http://www.fcsm.gov/working-papers/spwp22.html we find a working paper from the Federal Committee on Statistical Methodology, which has deeper details on actual operations.

From that page, the "Statistical Disclosure Limitation: A Primer" document has an interesting section defining inferential disclosure - "occurs when individual information can be inferred with high confidence from statistical properties of the released data."

And the "Current Federal Statistical Agency Practices" describes the multi-dimensional linear programming used to prevent that, along with other techniques including geographic thresholds, population thresholds and coarsening.

So the summary is: Yes, it is a serious issue to be concerned about, but Census is taking it seriously, applying some real science and math to it, and it looks like they are doing a good job.

SIAM (Society for Industrial and Applied Mathematics) has several special interest groups related to computing/programming problems. The other major math and stat groups have excellent articles on computing problems from time to time as well like the AMS, MAA, or AMSTAT, but SIAM probably provides the most of these groups and a lot of coverage that compliments IEEE and ACM. Also depending on if you're working in a specific industry or if you're furthering your studies in a graduate program there may be other professional societies that deal with informatics or computational issues related to that focus.

I love all of these "reasons" for not using Excel that seem to boil down to "When I learned to program, we didn't have ones and zeros, just zeros, and were were glad to have 'em." I use Excel occasionally for fast, free-form calculations and even exploration, but never for real research. The reason is simple: it gives incorrect answers. Check out the the links here for details: http://scholar.google.com/scholar?hl=en&lr=&q=On+the+accuracy+of+statistical+procedures+in+Microsoft+Excel+97&btnG=Search This one is easy to read and gives a short, detailed list of some of the problems: http://www.amstat.org/sections%5CSRMS%5CProceedings%5Cy2001%5CProceed%5C00470.pdf

Not entirely the same, but still interesting: http://www.amstat.org/PUBLICATIONS/chance/103.myth 0.pdf. It's 10 years old though. Still, there are some relevant numbers in the article. Atlanta Police Department reports show that, in 198 cases of burglary, offenders obtained the victim's gun in 6 cases. Victims were able to use their gun in self-defense in only 3 cases.

You seem to have not paid attention. There were wide, and well substantiated claims of Diebold problems in that election, many of which are open to interpretation as to whether or not they were fraudulent, and if so if they were politically motivated (as opposed to just covering for incompetence). The problem was that the elections were so overwhelmingly in favor of the Democrats that these irregularities had little impact.

Here's the thing that bothers me about this, though: EVERYONE should be upset when someone says something like that. His statement should be read as, "I will use the Republican party to tear down democracy." Republicans should take that as a slap in the face, and should be MORE outraged than Democrats!

Pot/Kettle

Other than your attempts to divert the issue away from one of the most amazing statements ever to be made in the political world, I'm not sure what your point there is (BTW: I'll note that I haven't told you my party affiliation... just for fun, I'm going to continue to not do so, as I love watching people guess).

As an ofttimes Republican voter I was:

* Upset when a thorough study of the 2000 Florida results (by mostly liberal mainstream media) concluded that Bush had indeed won the state, but long after that history has been re-written to only focus on headlines from 24 hours after the election.

This is unrelated to my point, but it so happens it's wrong. In reality a number of news outlets worked together to determine which of the statistical methods that various parties were pushing for made sense, given the situation. The conclusion was not particularly aimed at a "who won" sort of result, but the findings did indicate that Al Gore's chosen method of recount would not have made a substantial difference in the outcome, while a state-wide recount most likely would have resulted in the opposite result.[1]

* Upset that in the 2004 election cycle, Democrats AGAIN cried foul in districts where lazy, incompetent election officials spent taxpayers money hand over fist too buy this junky hardware and software rather than do realistic requirements analysis, and particularly upset that while these changes took place (pre 2004) these same left wing complainers said NOTHING, waiting instead for the results that they didn't like.

This is not fair. I watched as plenty of "left wing complainers" screamed bloody murder in the non-mainstream press, but the mainstream press didn't want to touch the story because it was too "technical." Once the election was on... then it was news. Either way, the fault does not lie with the "complainers", but with those that were producing the bad systems and selling them to states that were hard-pressed to meet ill-interpreted new regulations.

* Post 2004 I engaged in numerous debates with individuals who swore they were not complaining about the 2004 results, but were indeed interested in the issue. Again, these people grew silent as distance from the 2004 election increased. As I would bring up newly discovered issues with this Windoze based software (and hardware) they would thank me politely for reminding them and then go right back to Bush bashing, continuing to ignore the very REAL bipartisan problems that were being ignored by the mainstream media.

Many of us have given up, and it's nice to see in this article that Diebold is finally deciding that they don't want to be in that business.

* While you say the results of 2006 were "overwhelming", in fact in many local elections the margin of victory, particularly for newly elected democrats were a dozen votes, and for state level offices a few hundred. Almost NONE of these votes were contested, even in cases where the Republican candidate had only to request a recount.

Of course. An overwhelming margin in the U.S. for (just for example) a Presidential race is what, 10-15%? That me

I did a rewrite of a similar study back in January 2002 for the American Statistical Association. [PDF version. Google's HTML cache.]

It found, essentially, the same thing. That study looked at accident records and found that the effect of cell phone usage was similar to borderline legal intoxication, but less than the level of impairment suffered by the typical person busted for DWI.

Ok, so according to this article (http://www.guardian.co.uk/US_election_race/Story/ 0,2763,410902,00.html) the ballots were supposed to be kept under lock and key and never counted. Which is obviously incorrect, as can be seen here (http://www.amstat.org/misc/PresidentialElectionBa llots.pdf), that Florida has a Sunshine law which allows all of its ballots to be examined. The guardian article had other errors.

Quote:
"The more immediate victim however, besides Al Gore and the system, is the supreme court itself. As the liberal lion of the bench, John Paul Stevens, put it in his strident, dissenting opinion: 'Although we may never know with complete certainty the identity of the winner of this year's presidential election, the identity of the loser is perfectly clear. It is the nation's confidence in the judge as an impartial guardian of the law.'"

He was talking about the USSC ruling undermining the Flordia Supreme Court ruling, making it seem like their ruling had been done for partisan reasons. The article makes it seem like he's making the opposite claim, which is deceptive. I've read over the Flordia Supreme Court decision and consider it very reasonable and based on the correct interpretation of the law, so I understand where Justice Stevens was coming from on this.

Here's the results of one newspaper recount:
http://www.pbs.org/newshour/media/media_watch/jan- june01/recount_4-3.html

Here's the results of another:
http://www.norc.uchicago.edu/fl/index.asp

Both found that Bush probably would have won if the recounts had gone through anyway (with margins of error, it's impossible to tell, of course). NORC was disingenuous in claiming that 4 methords would have given Bush the victory, and 4 Gore, when the Gore methods mainly involved determining intent on overcounts, which is illegal in every county in America. (I.e. if someone voted for both Buchanan and Gore, then claiming it is a vote for Gore. Sure, it's probably what happened, but it's still illegal.)

This paper is on the web and helped me to an amazing degree: "The Science of Scientific Writing" http://www.amstat.org/publications/jcgs/sci.pdf If you're looking for help on fixing little things: "Bugs in Writing"

Is Nature really a scientific journal? That's a serious question: it's not in my field, and I'm not familiar with it. As I recall, it's printed on glossy paper, and has advertisements?

I'm familiar with journals like Journal of Economic Literature (JEL) and American Economic Review (AER), which do have some paid staff, paid for through the American Economics Association dues and subscriptions. I'm also familiar with journals like the IEEE Spectrum and the American Statistics Association American Statistician (AMSTAT), which are glossy magazine with ads, and the Journal of the American Statistical Association (JASA), which has few or no ads.

Spectrum and AMSTAT are not considered serious journals in their fields, peer-reviewed or not, while AER and JASA are. The difference is that a non-Ph.D might read Spectrum or AMSTAT, but would not (probably could not) read JEL, AER or JASA. I think that Nature is in the same class as Spectrum, isn't it? It is intended for a broader audience than the few academics working in the field? It's not the primary journal of record for the discipline it covers? If so, I don't think that it's really the sort of thing this initiative is aimed at.

The problem isn't that Excel comes from MS or that it is closed source (except in a round-about way): the problem is that the flaws which make Excel unsuited for statistical analysis don't affect sales enough to justify the cost of fixing them.

I think that this picture says it all.

Pure foolishness. MS is all about sales. If the bugs didn't keep you from buying their crap all these years, why would they stop you now? Why would MS fix them now?

MS will never fix a bug unless that has a positive impact on their bottom line. This link will take you to a collection of info about bugs that MS hasn't fixed in Excell.

... 5 years later, 3 service pack later, windows 2000 *STILL* has that bug.

See above. They've got you by the Balmers, and they plan to squeeze you a good deal harder. THEY DON"T CARE IF YOU'RE HAPPY! They care about separating you from their money. By buying their product year after year, you've proven that they don't have to fix bugs in order to gouge you.

I have no idea whether OpenOffice suffers similar flaws. Perhaps I'll get ambitious enough to run some of these tests on it myself, someday. In the mean time, I do my number crunching in R. I find that LaTeX provides far more professional-looking reports than any combination of MS products.