I yam what I yam.

[elmark]

I guess it's about time to make a big long rambling livejournal post.

I'll separate this into possibly appropriate sub-sections.

I like what I likeOver the past week, I've discovered that I still like what I've liked in the recent past. And my recent past, I mean the past 10 years. Such a statement has not always been true. I've watched the first

Comments

[elmark]

My use of correlation of was inexact, but I think it's not too far off of reality. Connection with a person thinking Colbert is conservative and them being conservative, not that there's a particularly large probability that conservatives think Colbert is actually conservative/serious. Basically, "you hear/see what you want to hear/see".

Which is odd, because I feel that I typically see the opposite of what I want to hear or see, even when people are supposedly acting in a balanced fashion. Whaddyaknow.

At least I can say that age wasn't a significant variable in the study.

I can't divine from their numbers what their B numbers mean... I assume they're coefficients in a linear equation, but what variables and constants are, I dunno. I think I understand the p values... over .05 meaning statistically un-useful (It's directly related to the significance level). I guess I should look at the GPower docs before assuming that they're using familiar symbiology.

I do wish they'd published some raw numbers instead of just shoving the numbers into some stats software and pulling out the parts they want to talk about. I did like the fairly raw means and standard devs that were supplied, but what needs to be line-fitted here? I really don't see much reason for linear regression in this case, but I suppose I'm no social science guy. Not even a stats guy, really. (Oh the old short-lives dreams of being a probabilist.) Maybe I should look into the software they used (freeware, it is). It's not like I've had much of a chance to mess with least squares regression in the compsci field, even though Gauss is still me heeeero.

I should take a few days and paly with least squares regression. Just for the Gauss.

800k windows installer... that took... a long time? No. Such... tiny softwares. Mostly developed for MS-DOS... I'm scared. Sorta. Not really.

[challudym]

Ohh it's one of Mark's alternate personalities replying XP

It seems to me that trying to divine exact figures from a poll about how people 'felt' about something seems a bit useless. They could do everything exactly as they had a different day and get different results. The extent to which they express their opinion through their answers, and the way they interpret the question, is subjective, too. At best you could look at it and say, well, generally, they sort of tend to feel this way about it.
On the other hand, having fun is useless, too. Run free in your mathematical playground :D

[elmark]

This is why the physics students made fun of the social sciences back in the day. It's all good stuff to study, but some is far more inexact and likely to be oversimplified than any of the "hard" sciences.

[evanthered]

[elmark]

As I read it, I don't see that the anthropic principle has anything to do with my ranting or hard/squishy science. Yes, a physicist delved into the realm of theo-philosophical speculation, but this should hardly be considered as an action within the realm of modern physics. It is a fully interpretive theory, which predicts nothing that is not currently known and with no ability to be actively tested. The ability to reliably test and re-test predictions of a theory is what I hold as a tenet of "hard science".

Modern scientists which I have encountered in academia have a good record of keeping to the idea that their job is to find reproducible relationships between well established concepts such as masses and forces. The theoretical guys on the frontier have get a bit "out there" to come up with new ideas, but unsupported ideas are treated as just that... something that's nice to come up with a way to prove or disprove. I may have just been exposed to some really good teachers/researchers.

I probably was. Here are the pages for the three people that I think most influenced me as a maturing scientist... good teachers and responders to incessant questions:
http://people.bu.edu/dkcampbe/
http://www.hep.uiuc.edu/home/g-gollin/
http://physics.illinois.edu/people/profile.asp?serrede
It's not like I worked closely with any of these guys, but I heard them talk plenty and asked several questions.

The thing that physicists are having a hard time living down are things like aether theory, which was widely accepted until proven imaginary. This has prevented much bandwagoning in recent years, and led to a large quantity of "It's unknown, but there are some theories that might explain..."

You may note that I'm often the first person to say that any wacky physical theory is highly likely to be wrong. For instance, "wormholes", tachyons, and so forth. As an aside: I don't take the physics in sci-fi seriously (wondering/pondering the implications thereof), other than ways in which it should influence the plot. In most sci-fi shows, you shouldn't pay much attention to that, either, for maximum enjoyment.

We science-y guys are always real, though sometimes irrational.

You deserved that math joke.

[evanthered]

[evanthered]

[challudym]

To clarify, I wasn't describing the study itself as useless, just Mark's desire to know exactly what numbers went into and came out of the study.

[elmark]

My desire to know what numbers go into and come out of things is one of my greatest strengths. I plan to eventually build a career on it. :P