EPTUS Politics Mega-thread - Page 1086
| Forum Index > General Forum |
Now that we have a new thread, in order to ensure that this thread continues to meet TL standards and follows the proper guidelines, we will be enforcing the rules in the OP more strictly. Be sure to give them a complete and thorough read before posting! NOTE: When providing a source, please provide a very brief summary on what it's about and what purpose it adds to the discussion. The supporting statement should clearly explain why the subject is relevant and needs to be discussed. Please follow this rule especially for tweets. Your supporting statement should always come BEFORE you provide the source. If you have any questions, comments, concern, or feedback regarding the USPMT, then please use this thread: http://www.teamliquid.net/forum/website-feedback/510156-us-politics-thread | ||
|
Plansix
United States60190 Posts
| ||
|
Ghostcom
Denmark4781 Posts
On February 01 2019 06:19 Plansix wrote: I think it is productive for people to critically examine the construction of “hard vs soft” sciences and which ones are assumed to hold more merit. Because the construction is often used to discredit a specific branch field, rather than differentiate them in some meaningful way. I disagree. The softer sciences have terrifying issues with reproducibility which is obviously a massive flaw when trying to argue for principles being generally valid (which they do). | ||
|
Plansix
United States60190 Posts
On February 01 2019 06:24 Ghostcom wrote: I disagree. The softer sciences have terrifying issues with reproducibility which is obviously a massive flaw when trying to argue for principles being generally valid (which they do). Is that due to the flaws in the science itself or that the funding mechanism for those sciences favors novel results? Or even obtaining specific result? | ||
|
Ghostcom
Denmark4781 Posts
On February 01 2019 06:46 Plansix wrote: Is that due to the flaws in the science itself or that the funding mechanism for those sciences favors novel results? Or even obtaining specific result? Define "science itself". The other issues you mention are an issue in all sciences to some degree. | ||
|
Plansix
United States60190 Posts
On February 01 2019 06:51 Ghostcom wrote: Define "science itself". The other issues you mention are an issue in all sciences to some degree. I prefer to stick to the older definitions of sciences, physical, biological and psychological, rather than hard sciences vs soft sciences. They are just better defined and easy to explain. And the reason I asked the question is I wanted to know if the sciences themselves are bad; or are there political and monetary motivations behind creating flawed studies in specific fields? | ||
|
Ghostcom
Denmark4781 Posts
I think it is important to remember that just because a field is imperfect it can still hold plenty of merit. What I am trying to get at is that some fields which hold plenty of merit are so imperfect that currently they are not really achieving new insights but only obfuscating the picture. A specific example is psychology (I am obviously painting with very broad strokes here, but I would have to add too many disclaimers otherwise which would render my post entirely unreadable) which suffers massively from a lack of reproducibility. Which for a long time was somewhat wilfully ignored. Money, politics, and prestige all played a role in the wilful ignoring. | ||
|
Slydie
1900 Posts
But, no scientist exists in a vacuum, there are founders, colleagues, media, dedicated magazines, poiticicians etc. which are all extremely important deciding what is being researched and what is not. Becoming a researcher is very difficult most fields, and navigating this extremely competetive environment and finding the right project (that the establishment wants?) can be as important as being good. There have been some nasty cases of cutting corners and downright cheating revealed. One such lie was about authism and vaccines, and it still does tremendous damage, even after being exposed and disproven many times. | ||
|
TheTenthDoc
United States9561 Posts
It ends up wasting a lot of resources: drug companies that run prelim trials on large number of compounds routinely select the ones with largest p values and effects and are then shocked when in subsequent analyses they don't perform as well. Of course they won't, you picked the ones that were most impressive! In some ways it's like if people trying to estimate change in the speed of light across mediums had decided to just test "is the speed of light different in medium X than in the air." This is all ignoring the central lie of clinical equipoise that lies behind virtually all drug trials particularly for placebo-you would never bring something to market that doesn't have an advantage in at least one thing you're testing-but this will likely never be exposed because nobody feels more comfortable pretending their randomization prior is 50% harm/50% benefit. Edit for more wankery: + Show Spoiler + And for one additional issue, when people use p values as a benchmark for replication, even if an original study was correct and they're estimating the exact same thing there ought to be a 50% chance a first study with p value 0.05 will have a second p value over 0.05! | ||
Womwomwom
5930 Posts
On January 31 2019 15:46 iPlaY.NettleS wrote: How very authoritarian. And not something i can agree with. Many these days do not have the immune system to cope with vaccines.Heck just two weeks ago one of the top UK cancer researchers died minutes after recieving the yellow fever vaccine! https://www.theguardian.com/uk-news/2019/jan/11/top-cancer-scientist-prof-martin-gore-dies-after-rare-reaction-to-yellow-fever-vaccination A more moderate position on vaccines is required.Recognising that there are risks.Being on either extreme is not good If he was travelling to any of those regions, which is the only reason anyone would get a yellow fever vaccine, there's actually more chance of him contracting yellow fever (and probably dying from it) than him from suffering complications from the vaccine itself. If he suffered from complications from the vaccine (a weaker dose), there's no question that contracting yellow fever from a mosquito would have probably killed him. Not to mention that there's absolutely zero mention that he died minutes after getting the vaccine. The complete inability to understand statistics is something that anti-vaxxers/pro-choice advocates all have in common. The reduction of yellow fever cases in a lot of African countries is proof that the risks significantly outweigh the 1/400,000 (that's 0.0000025%) chance of severe complications. Are there risks? Yes. Do the risks outweigh the benefits? Not remotely close. The only people who don't get vaccinated are those who are immune system compromised. That's generally people with actual diagnosed autoimmune disorders, the extremely young or extremely old. Everyone else has no reason to not vaccinate. Even then, the risk of complications for the extremely old (>65) are like 2-5 per 100,000. If you're going to tell people to be dangerously careful with vaccines at those rates, you better start telling people to not swim or drive because you're probably more likely to drown or get into a car crash. | ||
|
Mohdoo
United States15401 Posts
On February 01 2019 07:00 Plansix wrote: I prefer to stick to the older definitions of sciences, physical, biological and psychological, rather than hard sciences vs soft sciences. They are just better defined and easy to explain. And the reason I asked the question is I wanted to know if the sciences themselves are bad; or are there political and monetary motivations behind creating flawed studies in specific fields? To me it is all about how the information is discovered/processed. If someone uses the scientific method, it is a "science" because of methodology and feedback loops. The whole idea of: 1. I think x is probably true 2. I'll design an experiment using statistics and stuff to see x is true 3. My data implies x is not true, but rather implies y is true 4. I'll design a new experiment aimed at confirming if y is true 5. Cool, looks like y is true Things like "I think increasing food stamps funding ends up allowing more kids to go to college" can be studied in that way, but it's not like we can measure a bond length or something the way we do chemistry. I think that justifies separating between "hard" and "soft" because the two types of science are conducted very differently. But they are both using the core scientific method of accumulating and filtering information. I get what you are saying about labeling something as "soft science" is sometimes used condescendingly, but I don't think that is the case here. I can't run a chemistry experiment to see how effective foodstamps as a program is. But we can still use the scientific method. | ||
|
Plansix
United States60190 Posts
Edit: to expand on that further, I would question the hypothetical study above’s merit simple because the food stamps program is design to make sure people have food. Drawing a direct line to college admission requires excluding to many extenuating factors to be useful. | ||
|
m4ini
4215 Posts
On January 31 2019 14:04 TheYango wrote: This is a bit of an oversimplification, and doesn't serve to help the argument. Wegandi's disagreement isn't with the basic science underlying climatology. His criticisms of the models for climate change e.g. Are real criticisms. There are plenty of imperfections in current climate models, and climate change deniers love to point these out as invalidating. Where Wegandi (and others) go wrong is when he says: Which belies a belief that a "proven" model needs to exist for policy to be made on climate change. That's not how science works, and not how science-related public policy works. His view belies a pre-modern understanding of science, before Bayesian statistics was a core element of literally every experimental science--when science was like math or philosophy and interested in "truth" and "laws" rather than "models" and "95% confidence intervals". That models for climate change have flaws and inaccuracies is real, but also not really relevant when it comes to applying those models and making policy. Proof and certainty are not requirements for the use of scientific models when making public policy because uncertainty and probability are baked into how those models work. What matters is the expected value of public policy decisions based on these models--with the likelihood of the model being right or wrong being a matter of Bayesian inference with current climate data as our prior. The "certainty" and "proof" of these models don't matter in and of themselves, they only matter insofar as they affect the likelihoods of various outcomes, and with them, the expected value of the aforementioned policy decisions. This is where many climate change deniers simply don't understand what the scientific consensus on climate change actually means. The scientific community certainly acknowledges the flaws and imperfections of current climate models. But consensus has more or less been reached on the fact that the models are "good enough" such that making public policy decisions based on them has long since passed the point of having a positive expected value. Even if they models end up being wrong, or even if making said public policy changes leads to a bad outcome, that does not mean that those changes should not be undertaken because the expected value of making those changes is good. It goes back to the coin-flipping analogy I made on the last page. The fact that you can't prove that the coin is weighted towards heads is irrelevant. It's still stupid to bet on anything other than heads based on your priors. The coin might actually be weighted toward tails, but based on your prior information, the likelihood of the coin being heads-weighted is higher than the probability of it being tails-weighted, so logical choice is still to bet on heads, and then adjust as you get more data. But climate change deniers and antivaxxers fixate on that one tails and insist that you should be betting on tails just because you can't prove that the coin is weighted towards heads. Sorry, that's simply untrue. I literally said, scientists might be wrong or not in regards to the effects of warming especially in regards to highly volatile systems like hurricanes etc. What i'm saying is that it doesn't matter fundamentally. Fundamentally, if you pump energy into a system, you'll change it. There's zero way around this, except, as i said, if you believe that laws of nature/physics as we currently know them are wrong. Which of course is idiotic to assume - you could say that they're incomplete for certain things, sure (gravity instantly springs to mind). But to argue that they're flatout wrong, where's the limit? Do we assume light speed isn't the universal speed limit either? Gravity is just imagination? To ignore some of the most fundamental and well proven concepts in physics is just idiotic, it's not creative or "forward thinking". We can start doing that when there's evidence (and by that i mean actual evidence) to the contrary. Which there is none. Yes, it's possible that climate scientists are wrong on the effect on hurricanes. Who cares? When the time comes and people start actually dying by the millions (when, not if), the only thing you achieved is that you're able to point out that it's not the hurricanes doing it, so scientists were wrong. It's stupid. And i mean, actually flawed or nonexistant knowledge of basic physics, or simply a bad faith argument knowing full well that you're basically bitching about semantics (as in, bitching about one of the symptoms maybe being wrongly predicted, when the result in the end still is annihilation). Put it this way. Make a campfire. Now take Wegandi and tell him that if he puts this piece of paper on top of the fire, and then his face right after, first the paper will burn and then his face gonna be crispy. There's no doubt for anyone that if you put your face into the fire, it'll baconate. Now you got Wegandi who puts the paper on top of the fire, and thanks to the heat, it actually just floats away instead of burning - and with that, he has now proven that fire doesn't burn shit. This is pretty much exactly what people are doing. Something isn't working as predicted, so the entire thing must be wrong - ignoring that some fundamentals don't change. Fire is still hot and will melt your face - and putting energy into a system will heat it up. The positive feedback loop in regards to polar ice is also not even remotely disputed - if we warm the earth just slightly, more ice is going to melt. If you melt the ice, more heat gets absorbed by the earth since a large part of what we usually reflect back into space is now absorbed. That in turn melts ice faster (we're not even going into methane possibly being released, that again only makes it worse, but the outcome would still be the same), making it less reflective again. Even if we had bulletproof evidence that this is already happening, these people would come at "intellekshuals" with the argument that look, it's snowing. It's cold. How can the earth be warming if we just had record low temperatures? Which, again, for a person with just a basic understanding of science is easily explainable (melting ice requires energy, by melting the ice you take the energy out of the system - earth first cools, then warms due to a limited supply of ice on the caps). You actually can't win this argument until we literally watch people die by the millions. edit: sidenote, i don't mean to be rude to you btw, in case this reads like an attack. | ||
|
Acrofales
Spain17853 Posts
On February 01 2019 05:41 Plansix wrote: Yes. Only one of those displines is intrested in solving the problem of people not trusting vaccines. Also, have you examined the idea that something must have science at the end of it for it to be meritorious to public policy? Political science is sort of a thing. But there is no legal science. Urban planning isn’t a “science” by most traditional metrics. The study of languages is not a science either. Of course urban planning is a science. Just because urban planners aren't scientists doesn't mean there's no scientists studying the phenomenon. The rather stuffy term for it is Planology, but Urban Design is also used. It's generally considered a subfield of geography, with a lot of social sciences mixed in. Legal science is also a thing. A friend of mine got her PhD in law, studying how international law was being implemented at the ICC with.a specific focus on the case of Charles Taylor. | ||
|
Acrofales
Spain17853 Posts
On February 01 2019 08:36 Plansix wrote: It always seems to come down to a short hand to dismiss the findings of a study or field, rather than just talking about the methodology of said study. Edit: to expand on that further, I would question the hypothetical study above’s merit simple because the food stamps program is design to make sure people have food. Drawing a direct line to college admission requires excluding to many extenuating factors to be useful. For someone who was just trumpeting the merits of soft science you seem to know remarkably little about their methodology. You seem to think they have never heard of confounding variables, and have no idea what to do with them. Now I am willing to grant you that a lot of studies willfully ignore confounding variables under the pressure to get their shit published, and that leads to a badly applied methodology and the problem of reproducability. But that doesn't mean the methodology doesn't exist. And obviously, if a study is done using a bad methodology, its results should be dismissed. That's why good scientific studies go to such lengths to show they have applied the correct methodology (correctly), and why most peer reviews are most critical in analysing this. | ||
|
Ryzel
United States520 Posts
My particular field of ABA (applied behavioral analysis) is not quite as rigorous since a key focus is on the applied aspect (meaning we don’t often have the luxury of exploring theoretical concepts in controlled environments because Jimmy is banging his head right now), but despite this there have been several types of case designs created with the intent of establishing experimental control in a variety of circumstances (e.g. multiple baseline, alternating treatment, changing criterion, etc.). In addition, a key focus is on clearly defined terms and methods so that they are replicable (albeit in different environments), the idea being that easily replicated studies in a wide variety of environments can provide pragmatic information on the effectiveness of different interventions. That being said, I can also attest that there is little to no incentive for most to go through the effort of a full experiment just for replication purposes; however there is almost always something unique to each experiment that is worth considering (i.e. video modeling with adolescent girls, video modeling with primarily Spanish-speaking individuals, video modeling with previous victims of child abuse, etc.). More effort could be placed in this regard, but overall this particular field has come a long way due to this empirical focus and is arguably the most responsible for our improving ability to treat developmental behavioral disorders like autism over the last 20 years. TLDR; science doesn’t need to be 100%-right/closed-room/super-controlled in order to demonstrate effectiveness and make a lasting positive impact on society. | ||
|
Mohdoo
United States15401 Posts
On February 01 2019 23:30 Ryzel wrote: As an aspiring behaviorist (arguably one of the “hardest” subset of the soft science of psychology), I can attest that the training on how to properly conduct experiments and draw conclusions from data is rigorous. Statistics is of course a mandatory undergraduate class for all psychological fields, and a wide variety of statistical tools (Cronbach’s Alpha is one example) have been created that help analyze qualitative data that is otherwise difficult to interpret. My particular field of ABA (applied behavioral analysis) is not quite as rigorous since a key focus is on the applied aspect (meaning we don’t often have the luxury of exploring theoretical concepts in controlled environments because Jimmy is banging his head right now), but despite this there have been several types of case designs created with the intent of establishing experimental control in a variety of circumstances (e.g. multiple baseline, alternating treatment, changing criterion, etc.). In addition, a key focus is on clearly defined terms and methods so that they are replicable (albeit in different environments), the idea being that easily replicated studies in a wide variety of environments can provide pragmatic information on the effectiveness of different interventions. That being said, I can also attest that there is little to no incentive for most to go through the effort of a full experiment just for replication purposes; however there is almost always something unique to each experiment that is worth considering (i.e. video modeling with adolescent girls, video modeling with primarily Spanish-speaking individuals, video modeling with previous victims of child abuse, etc.). More effort could be placed in this regard, but overall this particular field has come a long way due to this empirical focus and is arguably the most responsible for our improving ability to treat developmental behavioral disorders like autism over the last 20 years. TLDR; science doesn’t need to be 100%-right/closed-room/super-controlled in order to demonstrate effectiveness and make a lasting positive impact on society. Good post and I agree entirely. The core goal of science is to analyze, characterize, predict and improve the world around us. Behavioral psychology does that, they just have to go about things a little differently when compared to chemistry. My view is that so long as you apply the scientific method to analyzing, characterizing, predicting and improving the world, you are a scientist. | ||
|
Mohdoo
United States15401 Posts
Overall, I am fine with this video and I remain skeptical, but hopeful. His voting history isn't what I would have wanted, but Trump used to be a left of center democrat. Yet look at what Trump is doing now. People do change, but I think Cory Booker was likely "this version" the whole time, just making deals with the devil to get where he is. At the end of the day, if he is going to push for what I want, I don't care about his past. I see the only reason to care about the past is skepticism of sincerity. Winning a democratic primary won't be possible as anything less than a progressive. It is important to keep in mind Clinton only won the 2016 primary because of her insane headstart and the fact that no one ever questioned she'd be president. Bernie doing as well as he did against Clinton was a total David and Goliath moment, but Goliath was just too huge and the system was busted. After 4 years of Trump, the left is going to be super pissed and get tossed much further to the left. I'd say we are at a point where M4A (or something similar) as well as nationally legalized marijuana are the bare minimum for having a shot as the presidency in 2020. His website still doesn't have a list of various positions, but he does seem to be trying to milk the MLK thing pretty hard. I am not entirely sure what he would do about it other than police reform, though. Great place to start. I remain skeptical, but we'll see. | ||
|
Plansix
United States60190 Posts
| ||
|
farvacola
United States18819 Posts
| ||
|
Mohdoo
United States15401 Posts
On February 02 2019 02:22 farvacola wrote: I wanna see Booker directly address his state’s relationship with big pharma. Without that, I’m just as skeptical of him as Beto relative to energy. Yeah, that's fair. I suppose I am not even thinking about big pharma as an issue because any dem who sides with big pharma in 2020 may as well not even run. He has a choice between the two, and if he choose big pharma, there's no reason to even discuss him. Bernie and other progressives will blast him all over Reddit until he drops out. | ||
| ||
