Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:Three puzzles (Score 1) 180

by ceoyoyo (#49496861) Attached to: Social Science Journal 'Bans' Use of p-values

I assume you're getting at multiple comparisons because you said "he measures many things."

You're right, the researcher should correct his p-value for the multiple comparisons. Unfortunately, alternatives to p-values ALSO give misleading results if not corrected and, in general, are more difficult to correct quantitatively.

Comment: Re:Mis-use=reviewer don't do their job (Score 1) 180

by ceoyoyo (#49496837) Attached to: Social Science Journal 'Bans' Use of p-values

Your 95% confidence interval (roughly*) indicates an interval containing 95% of the probability. The p-value indicates how much probability lies within a cutoff region. What most people do with a 95% CI is look to see if it overlaps the null value (zero, or the mean of the other group, for example). The p-value gives the same information, except quantitatively.

* yes, Bayesians, technically the 95% credible interval, from a Bayesian analysis, contains the area of 95% probability. The confidence interval, technically, isn't quite the same thing. Practically, in the vast majority of cases, the two are either mathematically equivalent or equal to within a large number of decimal places.

Comment: Re:What's the problem? (Score 1) 180

by ceoyoyo (#49496035) Attached to: Social Science Journal 'Bans' Use of p-values

There really aren't any good ways to measure those other effects. If you knew how your experiment was biased, you'd try and fix it.

Criticisms of p-values usually fall into two groups. Some people believe that p-values are bad because some people interpret them as the false positive rate. Personally, I think that's a problem with some people, and not p-values. The other criticism, which is particularly prevalent in social sciences, epidemiology and some of the squishier medical-type areas, is that if you get a non-significant p-value you discard potentially useful results. The usual proposal (which is probably the situation in this case) is to use confidence intervals. That way you can see all the area where your confidence interval is not overlapping zero! I have two objections to that. First, CIs are simply calculated from p-values and vice versa - they're really the same thing presented differently. Second, the reason you discard your result (or save it for a meta-analysis) if you get an insignificant p-value is because your data has been ruled insufficient evidence. Looking at CIs and marvelling at all the potentially meaningful area between them is just softening the p 0.05 rule of thumb. Incidentally, the false positive rate people suggest doing the opposite - using p 0.01 or 0.001 as the threshold for significance.

Comment: Re:Students + Anonimity (Score 1) 231

by ceoyoyo (#49489491) Attached to: Can Online Reporting System Help Prevent Sexual Assaults On Campus?

That's a good way of putting it. It might actually be a decent tool for the cops to use. The difference being that the police and courts are (supposed to be) knowledgeable about the law, trained in its enforcement, and accountable for their actions. The university offices in charge of these things, not so much.

On the other hand, if you're not even willing to walk down to the campus police station and file a report, any prosecution probably isn't going to go very far anyway.

Comment: Re:all in the implementation (Score 1) 113

"The reason Boeing went for this was to reduce weight, power consumption and complexity."

No, it's not. They most certainly are not running the entertainment system on the same wires as the avionics. The avionics system is a real-time network that is different at a very low level. The FAA exception allowed Boeing to connect the two networks at a single point, using a "network extension device."

Comment: Re:Students + Anonimity (Score 1) 231

by ceoyoyo (#49487929) Attached to: Can Online Reporting System Help Prevent Sexual Assaults On Campus?

Recent events, such as the Rolling Stone farce, have shown that universities and the public are not responsible enough to act in the way you suggest. If the cops want to set up a rape reporting website, that's one thing, although I still think it's useless because a rape conviction should require physical evidence. Universities or random people doing this? That's just asking for a witch hunt.

Comment: Re:Students + Anonimity (Score 1) 231

by ceoyoyo (#49487893) Attached to: Can Online Reporting System Help Prevent Sexual Assaults On Campus?

That line of thinking is where a lot of the danger of this kind of thing comes in. Eyewitness testimony, never mind a "witness" clicking a button in an app, is incredibly unreliable. Rape is a serious crime, and should require physical evidence to convict.

The real solution to this problem is to eliminate the stigma behind reporting rape and having the proper evidence gathered. A bunch of reports of rape don't mean anything. A bunch of reports of rape with positive rape kits is evidence.

Comment: Re:all in the implementation (Score 1) 113

Yes, but that's a big if. If you've got physical separation then there's no if, and there doesn't seem to be a reason why the avionics network should be connected to anything else, and it's usually not. Boeing apparently asked for an exception to that rule. It would be interesting to find out why.

Comment: Re:Hmmm .... (Score 1) 113

You're making unfounded assumptions. Before they started putting those GPS displays in the seats I used to take my hand held GPS as carryon. I had all the same information and my Garmin was certainly not connected to the aircraft systems.

According to the slides somebody linked up above, the airplane avionics network is isolated from everything else (running a completely different protocol) except for a specific exception that Boeing got for some 777s, subject to a bunch of security requirements. If I were in charge of the FAA I wouldn't have granted that exception. So if you're paranoid, you might want to choose any other airliner instead of a 777.

You will have a head crash on your private pack.