Slashdot Mirror

← Back to Domains

Domain: americanresearchgroup.com

Stories and comments across the archive that link to americanresearchgroup.com.

Comments · 8

  1. Re:Solution by JoshuaZ · · Score: 1 · on Cities Are Scolding Countries at UN Climate Conference To Cut Emissions (vice.com)

    1018 polled is not 320 million, but 1018 polled is a very decent sample size. I'm not going to bother going through the basic math explaining how sample size to get a given margin of error doesn't grow linearly with the population size, so instead I'll just point you to this margin of error calculator http://americanresearchgroup.com/moe.html, where if you put in 320 million and 1018, you'll see you get a margin of error of around 3%. In practice, one should due to issues with polling expect a margin that is about 1% larger, so that's around a 4% error, close enough to get the conclusion in question. So what is your objection?

  2. Margin of error is vastly underestimated by Tony+Isaac · · Score: 2 · on Researchers Unable To Replicate Findings of Published Economics Studies (businessinsider.com)

    For most studies of any kind, the margin for error is around +/-3%. For example, a study covering the United States population using a sample size of 1000 will yield a margin of error of 3.1%.

    So a study says that texting while driving increases your risk of a fatal crash by 23 times. That sounds like a lot! But hold the phone...the overall rate of traffic fatalities is about 10 per 100,000 people, or about 0.01%. Multiply that by 23, and you get 0.23%. A big change, right? But that 0.23% is still well within the margin for error of most any study.

    I'm not saying that texting while driving isn't dangerous. I'm just saying it's a lot harder to prove a link than it would seem.

    Still, people are wowed by big multipliers, and news writers love to tout dramatic statistics, whether the subject of the study is economics, medicine, or traffic safety. But if you understand statistics, you know that most of these studies don't really tell us much. It's no wonder we keep getting contradictory study results!

  3. Re:No 3G and No Touchscreen Keyboard? by iamhassi · · Score: 1 · on Amazon Disables 3G Web Browsing For New 3G Kindle Touch

    It's likely they shoved out a hobbled $80 model with the expectation that most people will buy the next one up but they can still claim the lower price point.

    No.

    $199 Kindle Fire = Upgrade to $399 Kindle

    $79 Kindle = for people who never owned a Kindle before or used the free Kindle smartphone app.

    $79 is also a great price point for Christmas gifts. The average US adult spends $658 on gifts and $79 allows them to buy one for everyone on their list rather than only a few $199 Kindle Fires.

  4. Re:Debate the Republicans want by roystgnr · · Score: 4, Informative · on House Narrowly Avoids Having to Debate Impeachment of Cheney

    the people that want to impeach are significantly in the minority

    Where by "significantly in the minority", you mean 54 percent.

  5. Which is sad, since a majority favor impeachment by LarryWest42 · · Score: 2, Interesting · on House Narrowly Avoids Having to Debate Impeachment of Cheney

    As of July, a majority favored impeaching Cheney, and it was close on Bush:

    So the Congressional Democrats are distancing themselves from the majority and certainly from their party members. Probably partly because they get saturated with "info" from all the boot-licking media and consultants based in Washington and NY.

    Of course many people try to characterize impeachment as a fringe movement... because that's the only rhetorical angle left: per the Constitution or the will of the people, impeachment of Cheney is quite reasonable. Particularly considering the damage done to the country.

    Hopefully we'll see criminal proceedings at some point. There's a lot of personal responsibility that still needs to be apportioned.

  6. Re:If only it were true. by boybaha · · Score: 2, Informative · on Windows Loses Ground With Developers
    A sample size of 400 is actually a pretty good sample. Assuming a global developer population of 10,000,000, a sample size of 400 would give you a margin of error of 5%.

    With a population of the same size, in order to get a margin of error of 20%, you'd need a sample size of 25. If the North American developer population is even a third of this, you'd get around 5% margin of error as well, so I'd say that statistically speaking that 400 is a pretty good sample size.

    These calculations were taken from http://americanresearchgroup.com/moe.htmlAmerican Research Group's Margin of Error Calculator

  7. Re:Mandate to fight terror by sammy+baby · · Score: 5, Insightful · on The NSA Knows Who You've Called
    The desire of the vast majority of Americans to root out terror in the US has given the government the mandate to use communication records. The nefarious behavior of the government goes only as far as that mandate. If you want to rail against someone for the loss of privacy, rail against the great silent majority in America who will not tolerate a repeat of 911.


    Interesting. You've conflated the (obviously and unarguably true) fact that most Americans want the government to prevent terrorist attacks against us with the assertion that the administration is free to do whatever it wants in pursuit of that goal.

    Obviously, I disagree. Defense of our country still must take place within the framework of our system of laws and the Constitution of the United States. To the degree that the laws need amending, I think that they clearly should be - although the current administration has shied away from this path. Instead, the Attorney General has repeatedly asserted that laws governing the gathering of intelligence data, even domestically, are not within the purview of Congress to issue, and that the executive branch can simply disregard them. When Congress has offered to make changes to legislation to make it more palatable to the administration, their offers were rebuffed: simply put, the administration does not wish to be governed by laws, regardless of their actual content.

    As for the rhetorical device you use - that the opinions you hold are that of the "great silent majority" - I can only say that in polls on a similar issue (the "warrantless wiretap" question), the data would seem to hold otherwise. In a poll run by the American Research Group, there was a near 50-50 split on the issue of whether the president should be censured over the NSA warrantless wiretap issue.

    Republicans (33%): Favor censure: 29% Oppose censure: 57% Undecided: 14%
    Democrats (37%): Favor censure: 70% Oppose censure: 26% Undecided: 4%
    Independents (30%): Favor censure: 42% Oppose censure: 47% Undecided: 11%
    Total: Favor censure: 46% Oppose censure: 44% Undecided: 10%

      I assume for the sake of this arugment that if approximately half of those polled supported a censure resolution on this issue, then more than half would be opposed to the wiretaps generally.
  8. Re:How about empower the Electoral College by malachid69 · · Score: 1 · on How Would You Change U.S. Election Procedures?
    Yes, but....

    Using the Electoral Vote Calculator (http://www.americanresearchgroup.com/ev/) you can see that if CandidateA wins every state that has 15+ electoral votes and CandidateB wins the rest, CandidateA wins (discounting voter fraud, corruption, etc -- just using it as it is meant to be)... Thus: [I tried to do this as a UL, but Slashdot kept complaining about Short Line Length, sooo....]

    CandidateA wins CA(55),FL(27),GA(15),IL(21),MI(17),NJ(15),NY(31),N C(15),OH(20),PA(21),TX(34) for a total of 11 states and 271 votes

    CandidateB wins AL(9),AK(3),AZ(10),AR(6),CO(9),CT(7),DE(3),DC(3),H I(4),ID(4),IN(11),IA(7),KS(6),KY(8),LA 9),ME(4),MD(10),MA(12),MN(10),MS 6),MO(11),MT(3),NE(5),NV(5),NH(4),NM(5),ND(3),OK(7 ),OR(7),RI(4),SC(8),SD(3),TN(11),UT(5),VT(3),VA(13 ),WA(11),WV(5),WI(10),WY(3) for a total of 40 states and 267 votes.

    That means that if CandidateA got the 21.6% most populous states, then it wouldn't matter what the other 78.4% of the states thought. If you don't want to have the winner determined by which areas have the most population, then the Electoral College is not the right forum to fix this problem. It is also interesting that, although they were the top 11 populous states, they only had 57% of the people (per http://www.thegreenpapers.com/G04/ElectorAllocatio n.phtml?sort=Popu ).

    It is an interesting point. People complain that Sun has too much control over Java. They made the JCP, so their vote is exactly equal to that of Apache. Every member on the JCP Executive board gets 1 vote. If the national election were based on 1 vote per state, then CandidateB would win the above stats because he had almost 4/5ths of the states. As it is now, the candidate only has to get the top 11 populated states to win. I agree a direct election would not fix that problem -- but I don't think anyone can use the stats to show why the Electoral college protects the small states. The numbers above disprove that theory.