xmjx

421 reputation

bio	website
	location
	age
visits	member for	2 years, 9 months
	seen	Apr 21 at 16:35
stats	profile views	48

	421 reputation	bio	website		visits	member for	2 years, 9 months
37 badges		location			seen	Apr 21 at 16:35

summary answers questions tags badges favorites bounties reputation activity

31 Comments

suggestions reviews revisions comments badges posts accepts all

Mar 14	comment	P values seem to be the wrong way around @user1134241 The p value is not interpreted in a way like "low is good, great is bad". A low p means that it is unlikely to get an observation as the one you made or even more extreme (i.e. deviant from a given value) purely by chance. Period. If you compare your prediction and an observation, a low p value means that "it's unlikely to get this difference purely by chance - so maybe there is a systematic effect going on". A high p means, yeah, this (small) difference is likely due to chance, so don't bother and treat the two observations as resulting from the same distribution.
Mar 13	comment	P values seem to be the wrong way around In that case you want a large p value since that would imply that any difference is very likely due to chance. Beware, though, of using the word "to prove" in the context of statistics. To prove something is the privilege of mathematicians. Empiricists "show evidence for" their hypotheses: since the outcome of a random experiment is - well - random, it might not show the true value. This is also why e.g. the Cochrane Collaboration conducts meta analyses.
Feb 20	comment	When to use Fisher and Neyman-Pearson framework? Where have you read about that? Please, cite your sources.
Feb 12	comment	P values seem to be the wrong way around So, all in all, would you say I should redraw my attempt at an answer or is my write-up okay enough to stay online?
Feb 10	comment	P values seem to be the wrong way around @user1134241 No, r and p together are important. A correlation is merely a wild guess as to how two variables tend to change together. The r value is the measure of how much they change together and the p value is the measure of how certain we are about that. Having a huge r is awesome but if p is also great then the r value means nothing since it's quite probable (probability p) that such a correlation or greater occurred due to chance.
Jan 31	comment	What are essential rules for designing and producing plots? Even professional audiences will (I assume) months later only remember the steep line in the graph and will have forgotten the intercept and everything else. You can have all that data in the (flat) graph by labeling the extreme values correctly and still have the information in it that nothing actually changed oder time.
Dec 31	comment	Under what conditions should Likert scales be used as ordinal or interval data? Just for clarification: Nunnally/Bernstein suggest to treat a variable as continuous if it has at least 11 distinct values (p. 115). Where is that "12 points imply interval scale" rule of thumb from?
Dec 17	comment	How do I calculate projected figures for the next year based on past performance? Also you could tell your boss the story of the inductive turkey.
Oct 20	comment	Is my data distribution normal? (Tried Shapiro and Kolmogornov-Smirnov tests) Just wanted to say that you guys are really awesome. I learn lots of stuff from you even if my own answers are wrong and display a huge level of being underinformed. ;-)
Oct 19	comment	Is dynamically increasing the sample size okay, if stated a priori? Thanks everyone for your insight. I it were possible to accept multiple answers I'd do it :-(
Oct 19	comment	Is dynamically increasing the sample size okay, if stated a priori? I'll use that train of thought to get a starting point for @EpiGrad's power plot.
Oct 19	comment	Is dynamically increasing the sample size okay, if stated a priori? That's a really great idea. While I side with @Aniko's answer because it addresses my question directly, your suggestion is definitely more useful in the short term. I think it will take me some time to wrap my mind around sequential strategies.
Oct 19	comment	Is dynamically increasing the sample size okay, if stated a priori? Okay, that sounds interesting. I read about adaptive designs before but only on some slides so the paper reference you gave looks really useful.
Oct 9	comment	Are all ordinal attribute types nominal? The design of the object oriented program would be nicer.
Oct 8	comment	Why would all the tests for normality reject the null hypothesis? @dsimcha But the normal distribution is just a probability density function which could predict the number of observations in a given bin of the discrete variable. So, I would understand if you said "no real variable is exactly normally distributed and this is why normality tests will fail at some point". But for "discrete data cannot be normally distributed since it's not continuous" I'd like some reference. I'm really interested in that kind of stuff. Not wanting to start a fight here.
Oct 8	comment	Why would all the tests for normality reject the null hypothesis? @Henry I know. What I mean: choosing any normality test upfront has some probability of choosing one that will say "significant". So is it better to run a battery and then ... what? Average? Go with the majority vote?
Oct 7	comment	Why would all the tests for normality reject the null hypothesis? One more comment on the computer issue: please note that the mechanism often used to store decimal numbers in computers has a different granularity for the range of small numbers and the large numbers. So the minimum difference between to numbers that the computer is able to store is smaller for small numbers and larger for large numbers. For a computer, 100000.1 and 100000.2 might be the same while 0.1 and 0.2 are not. (Just an example - in the real world it's not that bad.)
Oct 7	comment	Why would all the tests for normality reject the null hypothesis? The Pearson chi square result looks like the data is not normally distributed. Just saying. What to do with such a result?
Oct 7	comment	Why would all the tests for normality reject the null hypothesis? Regarding everthing in the world being quantized: can't discrete data be normally distributed, too?
Oct 6	comment	Why is it a “statistical sin” to run a large number of correlations and only report the statistically significant ones? @DikranMarsupial As I see it, it's the "we do that once" approach that runs into difficulties as the result of the lack of repetition.