Slashdot Mirror
OOXML Vote and the CPI Corruption Index
Tapani Tarvainen writes "It turns out there's an interesting correlation between Transparency International's 'corruption perceptions index' and voting behavior in ISO's OOXML decision. Countries with a lower score (more corruption) on the 2006 CPI were more likely to vote in favor of OOXML, and those with a higher score were less likely. According to the analysis, 'This statistics supports with a P value of 0.07328 the hypothesis that the corrupted countries were more likely to vote for approval (one-tailed Fisher's Exact test). In other words, simplified a bit: the likelihood that there was no positive correlation between the corruption level and probability of an approval vote, that is, this is just a random effect, is about 7%.' Of course, correlation doesn't prove causality."
Brazil voted no, and we are a curruption paradise :)
Actually there was a strong fight in ABNT (brazilian standards regulation group) as happens in the government. The brazilian agencies and federal govern always used windows in a large scale from desktop to servers, while universities used most Unix and Linux. Current federal govern supports Linux, but there is a big resistance in sectors that always used windows and often there are problems with licitations (govern auctions) imposing a specific type of software.
For example, you can auction for "Microsoft Office" or "a Office Suite". Most states already fobid the first option in law, but at federal level it's not forbidden, even that there is a recommendation to avoid this kind of situation. But even with laws, often happens situations where the auction is so specific that it could only have a winner, when police does some investigation finds that someone got a "deal" to make sure company X wins.
Recently in my state they found a guy stealing mail stamps! Millions of them. He bought and kept the money for him, because the legislative house had a deal with the mail company to send letters. When found he buried the stamps in his house garden. Can you belive it?
Not only is 0.07 not significant, they used a 1-tailed test, rather than a 2-tailed test. If they had used the 2-tailed test, the p-value would have been 0.14, which is REALLY not significant. You're only ever justified in choosing the 1-tailed test over the 2-tailed one if you know for certain which way the influence is pushing. If, for example, one could make the case that the OOXML vote would have gone the other direction, with the more corrupt countries voting against it (a case we have no a priori reason to discard), then the use of a 1-tailed test is inappropriate here.
Actually, having read TFA, I'm pretty sure that correlation isn't appropriate at all here. The corruption scores are discrete, categorical values, rather than continuous values. This calls for nonparametric methods. Start with chi-square and move on from there. You can't do correlation with a straight face if your variables are discrete, since there's no guarantee that the "distance" in corruption between 2 and 3 is the same as the distance between 4 and 5.
Sweden has a low corruption index, but there is evidence of irregularities there. See, I just used evidence to trump the statistics in the article...
No, the article doesn't say that. It says "We found that more corrupted the country is, the more likely it was to vote for the unreserved acceptance of the OOXML standard proposal."
Good that you mention Sweden though. The "irregularities" you mention were that Microsoft Sweden offered bribes to close business partners to vote "yes" to accept a suggested standard SIS had carefully evaluated over months and decided was worthless.
Being bitter is drinking poison and hoping someone else will die
just because 78% of the 16-18 drink large amount of soda and 93% of the 16-18 year old go to school doesn't mean there is any correlation between the two... That's not a bogus statistics example, that's just an example on how bad people (you in this case) understand what correlation is...
Correlation would be: 85% of the kids 16-18 attending school drink large amounts of soda, whereas only 40% of those who do not attend school drink large amounts of soda. That is an example of correlation.
A good bogus example would be: People who wear suits to work have on average a higher income then people who wear work clothes, there is therefore a correlation between how nicely you dress to go to work and your salary. Therefore the way you dress to work has an impact on your salary.
Please note that the correlation in itself is not the bogus part of the example, the bogus part is the conclusion made by myself. Statistic themselves are rarely bogus, and if they are they can clearly be shown to be bogus, the conclusions drawn are the problematic part.
Maybe. :-D
Apple - iWork'08 under Tiger can import OOXML documents, but it does so in a roundabout way. It imports OOXML files into it's native format (.pages, etc), but it can't write OOXML. (Not to mention you have to export the document back to office format, rather than save as). TextEdit under Leopard is said to be able to read/write OOXML as well as ODF, but I can't confirm this. The iPhone is also said to be able to open OOXML as well.
Novell - Novell's OpenOffice is said to be able to handle OOXML (according to wikipedia), but I have not tried it myself. According to the same article, OO.o 2.3 (not Novell) is said to have an OOXML importer - so I guess that could cover Linux as well.
Palm - Datavitz DocumentsToGo
Java - No clue.
IP Issues - No clue
Easy to work with - Not my area of expertise.
7% is 7%. Labeling that "signficant" or "insignificant" doesn't change anything.
Not really. It comes from a rounding of 4.6%. The 4.6% (actually 100-95.4%) comes from 2 standard deviations, in a normal distribution.