Hacker News new | past | comments | ask | show | jobs | submit login
The British amateur who debunked the mathematics of happiness (2014) (theguardian.com)
243 points by IntronExon on Mar 4, 2018 | hide | past | web | favorite | 213 comments



My wife is working on a PhD in psychology (she intends to do clinical work) and after hearing about the mathematical "training" they receive in order to do research, I was astonished at the lack of rigour. One personal anecdote stands out. Myself being a professional statistician, she would sometimes come to me for help with her assignments. On one assignment, her professor had misunderstood one particular statistic and had given then a problem that wasn't well defined. I helped her craft a polite email to her professor explaining the ambiguity and asking for a resolution. The professor got extremely defensive and did not answer the question, and none of her colleagues realized that the question was incorrect either. And don't get me started on the conclusions they draw from the smallest of trials... /rant Sorry.


Related to your last point, the slow-burning controversy about P-hacking/Data Dredging:.

- https://en.wikipedia.org/wiki/Data_dredging - https://fivethirtyeight.com/features/science-isnt-broken/

One particular favorite (used in the above): https://projects.fivethirtyeight.com/p-hacking/


quick question, how well does pre-registration protect against p-hacking?


P-hacking can be done by analyzing the data in different ways until a significant effect is found, and then failing to report all of the analyses done. There is no universal way to analyze data (especially what collected data to include); there are choices that are made and this leads to a “garden of forking paths” where if you look around the garden enough noise alone will give you ‘signicance’. Preregistration makes this impossible by declaring your analysis before collecting data.


As a psychology undergrad, it's hard to be surprised at the lack of rigor when you realized a large part of the faculty does not understand what science is and therefore does not feel the need to believe in it. I'm still astonished that the entire subfield of psychodynamics is allowed to persist based on "case studies" ie phenomenology ie not science.

As a scientist should, I reserve the right to change my mind should the data or my understanding of it. I freely admit that I currently know little, but I'm still taken aback by the shortsightedness of some of my professors with regards to their... Epistemological relativism.


I hear that this is a joke, but it ought to be a real course:

https://hardsci.wordpress.com/2016/08/11/everything-is-fucke...


Fair enough, though as a computer scientist my statistics training was also awful. I actually tried several different stats professors' courses before I found one where I managed a barely passing grade. They all taught terribly and preferred solving meaningless integrals to teaching statistical techniques, applications, and intuition. /rant as well


"The second-century Stoic sage Epictetus argued that "Your will needn't be affected by an incident unless you let it". In other words, we can be masters and not victims of fate because what we believe our capability to be determines the strength of that capability."

That is absolutely not what Epictetus meant. In fact, like other Stoics, he taught his students a technique called negative visualization that is literally the opposite of positive thinking. The author couldn't have picked a worse example of an ancient precursor to positive psychology.


I have read almost all of the biographies and works covering Epictetus. Just want to confirm you are correct. Stoicism at it's true core was how to react to a life that was foretold. Most stoics believed their date was predetermined but now how they dealt with it. In context, Stoicism was the cure to being dealt a rough hand by "fate". It holds up so well that many modern therapies, such as CBT, can be directly connected. Now, we of course don't think probably believe in destiny here on HN (in general). Anyway just providing some color for how off the author was.

Stoics lived with death as a way to take the right course of action. E.g. if I knew this was my last day with my daughter how would I act? Could I claim to have made good choices or would I have regrets.


It’s interesting how positive psychology is coming under attack from all directions, and I think reflects many of the shortcomings that have been festering in the field.

From the very liberal / progressive side, Barbara Ehrenreich (author of “Nickel and Dimed”) has been making the case for the last 10 years that positive psychology is simply the last ditch effort of capitalists to “maintain morale” while Western living standards triumphantly continue their 5th decade in decline.

From a conservative side, a lot of religious leaders have made the case that positive psychology is simply an attempt by secularists/humanists to dethrone the notion of morals, suffering (a big thing in Christianity) etc. and put human pleasure as the highest of all goals.


> while Western living standards triumphantly continue their 5th decade in decline.

Are living standards really lower continually over the last 50 years? At least in the UK I think that absolute poverty is down over a shorter time, median household incomes at up significantly, life expectancy is up. What's continually gone down?


All of those things for white men, at least since 2008.

Which you can read either way (oy, I must have karma to burn?). On the one hand, the anti-capitalist left in the US has historically had a contingent with racial undertones. On the other hand, literally every other group's station was improved at least in part through government mechanisms (equal protection under law, civil rights, etc.) rather than market mechanisms.


As a pasty white man myself, life has been improving steadily since 2008. My income I think has doubled since then, I’ve lost about 20lbs of fat and I’ve picked up some cool new hobbies!

True, I did have to get a relevant degree and not be an idiot at work to advance, but life is pretty good overall.

Now if you mean that being a White Man by itself isn’t enough anymore to get you a cushy life style, then yes, it’s not. Welcome to the place everyone else has been in since forever.


There's a lot of people who weren't invited to the party.

The "able bodied, but aren't working number is close to an all time high. I believe it's higher than when women stayed at home while hubby worked?

I have never seen this amount of homelessness. I see daily. Im becoming too hardened to the suffering.

People used to have the privilege of having two shitty jobs. Now--those jobs are coveted, as careers, by hard working immigrants--that see nothing wrong with fitting six people in a 10 x 12 foot room.

Look at the jobs on Craigslist. Work with us for three hours a day, or on call? So many are just rediculious. Buy a four door, late model car, we approve of, and go into a slow debt death. Or, the amount of thieving bastards who know they can get away with terrible work conditions, and pay that makes showing up debatable.

Crime is down. Cameras are everywhere. People/business don't use cash like they used too. Surveillance is everywhere. And like always, the rich seem to get treated better by the justice system? Fees/fines have never been so high. Only the wealthy can afford to commit crimes? Oh, and they do run afoul of so many laws.

Look at all the young scholars who are living at home. Many living at a family home that will probally not be passed down because mom and dad will be forced into a reverse mortgage.

In tech--yes, the party is fierce. I do notice the new cars. I notice the carefree attitude. There's a liberal air of "Hay--it's just hard work? I'm just more clever than that guy?".

I hope it all lasts.

I guarantee so many of you will be right here wondering what the hell happened, on a old device? I think a new version of the punk movement will pop up. The hippie movement won't be the joke it currently is. Hell--the denial might protect a lot of you when the money dries up?


2008 is a decade, not five.

But more importantly, what statistics do you actually have for this?


By the way, it turns out people have been fucking all along and that there actually aren't any black or white people. Never were. Someone made the whole thing up and it got a wee bit out of hand. Thought you might like to know, as it might save you an awful lot of time and worry.


What is the claim you are trying to make? I don't see any sarcastic "/s" present so I assume you are being serious.

Are you saying that there are none to little genetic/biological differences between different races?


I think he's saying that "races" are a made up category and theres just people. I happen to agree, there is nothing special about some gene pool with an evocative label.


Sure, but I don't think anyone has claimed otherwise? Obviously there's very little genetic basis for the concept of race as a big important thing, but social constructs are absolutely real even if they're a product of our collective imagination.


I'm having a little discussion on a parallel thread where someone absolutely is advocating for some sort of racial determinism and thats the context for the comment. [0]

[0] https://news.ycombinator.com/item?id=16517930


To believe that you would have to ignore the current scientific evidence which is becoming stronger and stronger with the acceleration in our understanding of the human genome.

Even without any of that evidence though, we still have books like the bell curve which people have tried hard to debunk but so far the key points made in the book about the genetic existence of race have not even come close to being debunked.


> current scientific evidence

there is no evidence for the existence of a race that does not rely on referencing the idea of a race. the idea is self-referential. to understand this, you must understand philosophy.

another way of saying this is that assigning a person to a race doesn't give us any more information than the information we used to classify him in the first place.

> so far the key points made in the book about the genetic existence of race have not even come close to being debunked.

ok, I'll bite: what defines a race? what test do I apply to a specimen in order to sort him/her into a race category?


> there is no evidence for the existence of a race that does not rely on referencing the idea of a race. the idea is self-referential. to understand this, you must understand philosophy.

Huh? Race is defined by a combination of genetics in the form of phenotypes and culture. I don't see the circular paradox here.

> another way of saying this is that assigning a person to a race doesn't give us any more information than the information we used to classify him in the first place.

But we can? We can say that Chinese are more likely to be shorter than africans. That asians are more likely to have black hair than blonde hair or that africans are more likely to have higher testosterone than european whites or indians.

Ask any doctor or pharmacist and they can list you a few ways they are taught about how to treat people of different races differently because of biological differences.

We can even go further and say for instance that ashkenazi jews have a higher average IQ compared to Africans due to a large part their genetic differences and not purely environmental reasons.

The bell curve which was published a few decades ago proves many of these kind of claims so I don't know where you get this idea that we cannot determine anything useful from race.

> ok, I'll bite: what defines a race? what test do I apply to a specimen in order to sort him/her into a race category?

I gave a definition previously which is pretty accepted, for how you test for a race the answer is that's pretty easy, 99% of the time you can just ask someone and they will tell you.

We can prove this through cluster analysis, where you take a sample set of people and first ask them what race they are from a set number of selections. They then can read SNPs from each of the participants and hand that to a computer without the self reported data the people provided.

What you see is that even with 100 random SNP's you have a near perfect correlation between what people say their race is and the computer correctly grouping them into the race they chose.

If you want to see some sources, here is an article that links off and summarizes the studies I am referencing http://thealternativehypothesis.org/index.php/2016/04/15/329...


> We can say that Chinese are more likely to be shorter than africans.

whats a chinese? whats an african? and what is it about those categories that makes it necessary for you to speak in probabilities and not certainties? if those categories accurately described features of reality, and were not mere artifacts that stem from your worldview, you would be able to rigorously tell me a relationship between gene(s) and features.

> Race is defined by a combination of genetics in the form of phenotypes and culture.

> for how you test for a race the answer is that's pretty easy, 99% of the time you can just ask someone and they will tell you.

exactly, its a made up category, i.e. not a real thing.

> We can prove this through cluster analysis, where you take a sample set of people and first ask them what race they are from a set number of selections. They then can read SNPs from each of the participants and hand that to a computer without the self reported data the people provided.

you can do the same thing for things like "hippie", "biker", "nerd" etc. it doesn't make those categories leave the intersubjective realm.


> necessary for you to speak in probabilities and not certainties?

You're replying to a person who is specifically referencing group stats and not in a deterministic way.

Here's one extreme statement:

- There is no such thing as race, in that E(X | race) = E(X) for all X that we care about

Here's another extreme statement:

- Race biologically determines your fate, e.g. E(X | race + C) = E(X | race) for any 'other information' C

Both statements as wrong. As usual, the truth is somewhere in between.

Even if you don't think that 'race' is the most salient metric, it doesn't help that our culture sometimes seems to willfully conflate culture, religion, ethnicity, and race in the most uncharitable possible way.

Eg. if you say something bad about Islam itself as a religion, one often finds themselves accused of implicit racism—and this is accusation is determined by seeing that one group of people of a certain 'race' seems to follow that religion more, so if you criticize a religion, you must be intending to attack that group of people.


> do the same thing for things like "hippie", "biker", "nerd"

while race is not the correct term to use there, breeding two nerds or two biker isn't likely to produce another nerd or biker, while the result of breeding two people of a common ancestry has a likely outcome.

> exactly, its a made up category, i.e. not a real thing.

race is a made up category, but you can't just hand wave all of it:

> what is it about those categories that makes it necessary for you to speak in probabilities [..] you would be able to rigorously tell me a relationship between gene(s) and features.

only because something hasn't been explored in its entirety doesn't make it less convincing of an argument. beside, genetic is rooted in statistics, because of recessive/dominant expressions of traits, so you have to talk in probabilities.


> breeding two nerds or two biker isn't likely to produce another nerd or biker, while the result of breeding two people of a common ancestry has a likely outcome.

this is preposterous on many levels. 1. the outcome of any breeding project can be described in terms of species, individuals, and traits. race is an unnecessary and poorly defined category. 2. the offspring of nerds and bikers have a non-trivial likelihood of turning out to be nerds and bikers. 3. the person I replied to wants culture to be a component of race.

> you can't just hand wave all of it:

It's well within the bounds of proper discourse for me to be a racial noncognitivist because I believe the concept to be incoherent, and protest its usage by requiring my interlocutor to tell me what exactly they mean by this particular nonsense term.

>beside, genetic is rooted in statistics, because of recessive/dominant expressions of traits, so you have to talk in probabilities.

look, theres a value to studying the interaction of genes and their expression in an organism. there isn't value in delineating a gene pool into "races" because the category is net harmful to understanding.


> whats a chinese? whats an african? and what is it about those categories that makes it necessary for you to speak in probabilities and not certainties?

I already answered that race is generally defined as a combination of genetics and culture. I don't see what your argument is, some kind of no true scotsman where because the definition of race isn't some set in stone definition that you can perfect attain with a genetic test it's completely useless as a grouping mechanic without something else to replace it.

The reason i'm speaking in probabilities is because that is what the facts are, that we have more variance within races than between. Which is to say there is no reason you can't have an african who is the shortest person in the world, or a chinese man could be the tallest just because the mean bears out the opposite.

> if those categories accurately described features of reality, and were not mere artifacts that stem from your worldview, you would be able to rigorously tell me a relationship between gene(s) and features.

You are demanding a standard of evidence that is beyond reason, do you not believe in evolution because there are gaps in the fossil knowledge? There is no reason you require to know every specific gene that influences height to determine that africans are genetically predisposed to be taller.

Not only can we deduce that genes play a predictive role in attributes like height and IQ through twin studies, but we have already begun to find genes that can predict physical attributes like height.

> exactly, its a made up category, i.e. not a real thing.

Category theory is made up, does that make it not real? Something that has no value whatsoever?

I really don't understand the argument, because you seem to believe race is subjective (even though we can prove it isn't based on what i previously said regarding cluster analysis) to the point it has 0 predictive value. Is that your argument?

> you can do the same thing for things like "hippie", "biker", "nerd" etc. it doesn't make those categories leave the intersubjective realm.

And guess what, we can make predictions based on this knowledge. We can say that a self proclaimed hippie is more likely to smoke weed than a self proclaimed Muslim or Christian. Do you believe we cannot do this?

Because if you believe we can do this with subjective social groups, I would say again based on the cluster analysis work i linked to previously, we can say that what people see as race almost perfectly corresponds to what a computer would sort them into based on purely genetics.

Race isn't some wavy concept where you have asians thinking of themselves as africans, or japanese thinking they are chinese. Sure you can point to exceptions but they are just that, not even 1% of the people in these studies when using more than just a dozen SNPs


The average "African" is not "predisposed to be taller" that is just a stereotype. Mainland Chinese people tend to be taller actually than the average West African.

Source: https://en.wikipedia.org/wiki/List_of_average_human_height_w...

African Americans on average are taller, but that is probably due to environmental differences and selection bias (slavery etc).

I would suspect mainland Chinese people are much taller than they used to be due to improvements to nutrition.


What are you talking about? Why is it then we can see that african americans are taller than asian americans? I would be pretty shocked if it turned out asian americans were more malnutrinted than african americans.

> African Americans on average are taller, but that is probably due to environmental differences and selection bias (slavery etc).

The papers I have seen don't really bear this out at all. Here is what the NHS says about it https://ghr.nlm.nih.gov/primer/traits/height

Right on the first line

> Scientists estimate that about 80 percent of an individual’s height is determined by the DNA sequence variants they have inherited,

That seems to back up what I'm saying.


I said African Americans are taller and that might be due to selection pressure (DNA) as well as environment (better diet, etc). My comment was not about Asian Americans at all. I was talking about Mainland Chinese in China, comparing Mainland Chinese to Africans in Africa given height data.

The comment I was responding to was that Africans are predisposed to being tall genetically, and my response to that is we do not have evidence of that given the data we have about these populations.


> The reason i'm speaking in probabilities is because that is what the facts are, that we have more variance within races than between.

thank you, exactly.

> africans are genetically predisposed to be taller.

"african" is not a category that has any meaningful genetic definition.

> genes play a predictive role in attributes like height and IQ through twin studies, but we have already begun to find genes that can predict physical attributes like height.

none of which means that "race" has any meaningful existence.

>Category theory is made up, does that make it not real? Something that has no value whatsoever?

it has value in organizing information. another system of organizing information might be harmful. suppose you had a system of organizing information that weighted irrelevant data and disregarded relevant data. would it have positive value or negative value?

> you seem to believe race is subjective

race is intersubjective, as is language and culture.

> to the point it has 0 predictive value

what are you trying to predict? either the trait that you select for when you delineate a sub-population is connected to the variable you predict, in that case you define the category by the specific trait, not some giant bag the size of a continent; or the trait you select for when you delineate the sub-population IS NOT connected to the variable you predict, so you're spouting nonsense. there is no third option.

> we can say that what people see as race almost perfectly corresponds to what a computer would sort them into based on purely genetics.

the computer is just doing what the programmer told it to do based on a bunch of genetic categories that are artificially constructed. you get a guy's genes. what makes you say that guy is asian? oh thats right, a hand wavy mix of phenotypes and culture. thats entirely objective, dont know what I was thinking.

> We can say that a self proclaimed hippie is more likely to smoke weed than a self proclaimed Muslim or Christian. Do you believe we cannot do this?

you're going to classify him as a hippie because he smokes weed and predict that he smokes weed based on the fact that he is a hippie. and you dont see the circularity?

> Race isn't some wavy concept where you have asians thinking of themselves as africans

actually it is much more fluid than you seem to think. parts of africa and asia are geographically close, and have mingled for millennia.

> Sure you can point to exceptions but they are just that, not even 1% of the people in these studies when using more than just a dozen SNPs

thats more an artifact of the limitations of your studies, if you know anything at all about this subject you should know just how genetically diverse the human species is.


> "african" is not a category that has any meaningful genetic definition.

If you want me to be specific, the negroid, is that specific enough for you? I assume not.

> it has value in organizing information. another system of organizing information might be harmful. suppose you had a system of organizing information that weighted irrelevant data and disregarded relevant data. would it have positive value or negative value?

Again, you haven't proved that race can't predict. You haven't even tried to disprove what im saying, you just keep banging on about this stupid aragument that "well what is the color orange? Is it red or is it yellow? or gold even?".

Should we throw away colors because someone might see a yellow and another sees gold? Are colors useless now?

> the computer is just doing what the programmer told it to do based on a bunch of genetic categories that are artificially constructed. you get a guy's genes. what makes you say that guy is asian? oh thats right, a hand wavy mix of phenotypes and culture. thats entirely objective, dont know what I was thinking.

So you agree that cluster analysis proves race can be objectively measured? Thanks.

> you're going to classify him as a hippie because he smokes weed and predict that he smokes weed based on the fact that he is a hippie. and you dont see the circularity?

No see you are just strawmanning, you know damn well I said self reported hippie and i was making a point about self reporting vs actions.

> actually it is much more fluid than you seem to think. parts of africa and asia are geographically close, and have mingled for millennia.

Ok but you agreed with the cluster analysis work i linked to that shows we can still put them into nice little groups with around 99% correctness so I don't see what legs you have left to stand on here.

> thats more an artifact of the limitations of your studies, if you know anything at all about this subject you should know just how genetically diverse the human species is.

The more SNP's the studies add the more clear the race picture is, how on earth could you get the opposite picture from the studies?

Are you even reading the studies or do you just assume they say what you believe? Because at this point your just strawmanning over and over and it's just not a productive conversation.


> the negroid

also lacking a meaningful definitinion.

> Again, you haven't proved that race can't predict.

actually I demonstrated that you are either using the trait you select for to predict itself (circular) or you use a trait to predict something its not related to (nonsense). there is no third option.

> So you agree that cluster analysis proves race can be objectively measured? Thanks.

you don't understand how that is impossible?

> self reported hippie

exactly, not a feature of objective reality.

> The more SNP's the studies add the more clear the race picture is, how on earth could you get the opposite picture from the studies?

because I understand the a priori nature of the modeling process.


> but so far the key points made in the book about the genetic existence of race have not even come close to being debunked.

Some would strongly disagree: https://www.theguardian.com/news/2018/mar/02/the-unwelcome-r...


I started reading that article a couple days ago and it's part of the reason why this subject as sparked with me. His article does a pretty pathetic job of trying to actually disprove anything about the science.

The way he smears Charles murray as if hes some closet white supremacists is just absurd if you actually have heard him speak or read his works.

Charles is one of the best example of someone who end up stumbling into the question on race and gets put on a fiery cross simply for not towing the line that race is purely a social construct.

It's a long read so I don't think you expect anyone to try and debunk it on the spot, so all i would suggest is people interested listen to the episode of Sam Harris's podcast with Charles and determine for yourself if Charles is just some shifty white supremacist neo nazi posing as a liberal or a scientist just trying to get to the bottom of uncomfortable truths.

https://samharris.org/podcasts/forbidden-knowledge/

If you really want a point by point rebut though, I know someone who did a video a couple days ago on the article https://www.youtube.com/watch?v=IxCKQupwZQM


I listened to the audio for a bit but gave up after coming across multiple logical fallacies.

The fundamental point, though, is that given the large variability of both genetics and IQ (or most other traits, really) within those groups commonly called "races", and big overlap between them, the group label itself is almost always of little use.


I'm saying I can invent races on the fly, then find genetic differences that would appear to back me up. Is like playing numerology with a large enough book, you can always find patterns. Finding patterns however doesn't mean that black and white as racial qualifiers are anything other than a highly inaccurate classification system that is used to align skin colour with social class, and which doesn't even fall within the gamut of actual skin tone. Is patently obvious where black and white came from as concepts, if you look through history rather than biology.

While I do not expect to win this argument on nomenclature within the next fifty years, I do hold out some hope over the next century.

edit - Also, am British, the use of a sarcasm tag is offensive to our people. Probably for genetic reasons.


I think you need to read up on some the cluster analysis work that has been done over the last decade. Your statement about inventing races and assigning them to people isn't true.

I linked this in another comment but ill mention it again, this article makes a comprehensive argument for race while linking off to the supporting papers instead of just me shotgunning a bunch of papers at you.

http://thealternativehypothesis.org/index.php/2016/04/15/329...

Look under the heading "Genetic Clusteredness" and you find a summary of papers such as Bamshad 2003 which explain how they can conclude a strong correlation between self reported races and objective measured genetics via SNPs.


I hereby invent the race classifier of tongue-dexterity. Society consists of those who can and cannot roll their tongue and there is a high priesthood of the folks who can bend their tongue into a 'W'. It is just as arbitrary and capricious, but has at least one benefit I can think of over classifying people by latitude adaptation.


> Are you saying that there are none to little genetic/biological differences between different races?

That's pretty much what any geneticist says.


Unless you go to china, and suddenly we find pretty much 0 geneticists who would claim that races are just some arbitrary construct with no predictive value.

To roll it back to the west though, I have seen surveys like this of anthropologists which show no con:

https://sci-hub.hk/10.1111/j.1548-1433.2009.01076.x

Or this paper, in particular page 313 which show biologists with a 70% agreement with the question of “There are Biological Races Within the Species Homo Sapiens;”.

http://sci-hub.hk/10.1002/tea.3660290308



I linked these papers to someone else but it seems like you could use them. Both show absolutely no consensus by anthropologists on race and even a general acceptance of race among biologists:

https://sci-hub.hk/10.1111/j.1548-1433.2009.01076.x

http://sci-hub.hk/10.1002/tea.3660290308


The statement on race by that group was adopted in 1998, and is based on work from 1996-1997. That's at least 20 years of possible advancement in the field, unaccounted for.

'The paper above was adopted by the AAA Executive Board on May 17, 1998, as an official statement of AAA's position on "race."'


Ha, more people live better now, both absolute numbers and percentage of global population, than ever before in human history.


Tell that to the crackhead living in the rustbelt with no job. I know you can make the stats look like living standards are on the up and up but you would be ignoring the huge issue of income inequality which is root cause of the political instability the west is experiencing.

The OP's point in this context is spot on, positive psychology not only has nothing to offer this growing lower class but it's more and more looking like it's a net negative to them.


Just because income inequality is up from the 50's doesn't mean people are worse off.

For $70 a month an entire family can have access to the internet, Netflix and Spotify.

Food is cheaper, more plentiful and comes in more variety now.

Healthcare is much better and more people have access to more treatments than ever before.

You really could go on and on. Even if real wages have stagnated for the working class, modern technology and the convenience that comes with it has made life soooo much better for the large majority of people. The can buy more with their wages, stagnant or not.


> Food is cheaper, more plentiful and comes in more variety now.

There is more variety, with more of a spread of quality over price points. Lower quality food is cheaper, higher quality food is less accessible.

> Healthcare is much better and more people have access to more treatments than ever before.

Healthcare is better if you can get it. Yes, there are more treatments available, for some, but there are more without access.


> Tell that to the crackhead living in the rustbelt with no job.

People addicted to drugs or without jobs are not new issues.

> I know you can make the stats look like living standards are on the up and up

That's a bit unreasonable, suggesting any and all statistics that disagree with your view are not trustworthy.

> you would be ignoring the huge issue of income inequality which is root cause of the political instability the west is experiencing.

Inequality and living standards are significantly different things.

Do you have statistics to support the 5 decades of falling living standards claim?


Side note: TIL that Barbara Ehrenreich has a Ph.D in cellular immunology from Rockefeller University.


The "Amateur" is a PhD student now and has lot of citations on his google scholar. I am happy for him to have such a good impact on the field even though he started late.

https://scholar.google.com/citations?user=feMcJ4UAAAAJ&hl=en


It seems he has found his niche: finding flags in Psychology papers. His most recent publications are refutations of other papers:

* No Evidence That Twitter Language Reliably Predicts Heart Disease: A Reanalysis of Eichstaedt et al.(2015a)

* Through a womb, darkly: Methodological problems in a recent study of fetal visual perception

* Emodiversity: Robust predictor of outcomes or statistical artifact?


Sounds like he's directly addressing the lack of replication that we bemoan so much.


Sigh.

"British semi-retired network engineer educated in mathematics at Cambridge finds mathematical flaw in popular psychology paper and enlists Alan Sokal's help to debunk it."

Sure wish folks would quit sucking on the "amateur vs. the establishment" teat.

That aside, this was a fun read, and I think it's another example of why scientific progress in the future is going to have to be more multidisciplinary than it has been in the past.


Educated in Engineering and CS at Cambridge, courses which employ far less intimidating mathematics than the Maths Tripos for which Cambridge is renowned. I think Brown could reasonably claim to be an ameteur Mathematician - and certainly an ameteur Psychologist.


Brown here... I dropped out of Engineering at least in part because the maths was too hard, and in CS I skipped several of the more maths-based classes and concentrated on programming, operating systems, and networks. Basically my maths never progressed a good high school level, and by 2011 it had certainly taken some steps back. I don't even consider myself to have the standard of an amateur mathematician. (It's only since I got into psychology, and hence regression, that I have learned what an eigenvalue is.)


I think it means he was an amateur at psychology, which does make it appropriate.


Sure, but he didn't find a flaw in the psychology presented in the paper, he found a flaw in the mathematics used in the paper, which he understood better than the paper's author(s).

This isn't to detract from what he did -- I'm glad he caught it and that it appears he's since made a hobby of catching others.

The problem, as other supporting replies to my comment point out, is that the notion that hobbyists working outside of academia can defeat established science has recently become a political tire fire, and this otherwise excellent article is being framed to add fuel to that fire.

It used to be that this was sort of restricted to young Earth creationists and other people that you could sort of roll your eyes at and ignore, but now this attitude has been attacking climate science and energy policy and environmental science, and now we've got an administration that has put people in positions to influence science policy in this country on the basis of politics alone. Now it's not so funny anymore.

The takeaway from the article should be on the increasing importance of a multidisciplinary approach in the sciences -- something that data geeks at HN should enthusiastically support -- and not this "amateur defeats scientists" hogwash.


> Sure wish folks would quit sucking on the "amateur vs. the establishment" teat.

I think it's an awesome narrative. What do you dislike about it?


The problem I have with it is that it helps to legitimize the view that an uneducated person's opinions and intuitions are more valid than expertise. In the current political climate, it's dangerous to continue stoking that narrative. As the parent said, the guy was hardly an amateur, but an antivaxer/global warming denier/etc. will view it as confirmation that establishment science is corrupt.


> it's dangerous to continue stoking that narrative.

What, even in cases where it is right? You are steering dangerously close to saying "We the Elite should close ranks". And that sort of thing is the fuel of the very fire that you are trying to put out.

The fact is as humans we have no option but to trust our own judgement on things: either by studying them directly or by judging which experts to trust. And this article shows you can't assume experts are right just because society puts them in a particular position.

In the article we told that Nick Brown, an IT guy who knew high-school level maths, who discovered that an important paper by well respected psychology prof was BS when it was presented to him at a university psych class.

Why had the profession swallowed it? Apparently because their maths is so bad that they just swallowed what they were told, and (according to The Guardian) the few with qualms didn't feel anyone would be interested in hearing the contrary view.


I don't think it's necessarily that everyone in the whole psychology profession has worse than highschool level math skills. I think it's more just that, at some level, you have to trust that your colleagues have done their work correctly. You focus on the important parts of the paper (ie. the hypothesis, the method, and the results) because there's not enough time in your life to give full scrutiny to every detail of every paper you read. So sometimes, something slips through the cracks, until it's caught.


At which level in the sciences are you ever supposed to trust that your colleagues have done their work correctly? That sounds more like the opposite of what you are supposed to be doing.


At the level where it takes more than one lifetime's research to reinvent your field from ground zero, it's literally impossible to progress the field without taking something on faith.

Well before that, though, it becomes impractical to re-check every single thing. And the longer-established a field is, the more you have to take on faith.


So then you have to go to other people in that field and have them look at it. Sure, you have to take them on faith too, but a little bit less each time unless they all wildly disagree, and they can explain their reasoning to you in layman terms when you canvas their opinion. Now, none of this guarantees success, but it is a long way from trusting others to just do their work properly.


At the level where you cite the work of others.


I'm reminded of the Russian proverb; "Trust, but verify".

A citation does generally mean that you hold a given paper with a certain level of trust, but that isn't the same as having trusted the people who wrote it to have done their job properly.

The trust in the paper could be because you trust the folk who wrote it to have gone and done their job properly, however the trust could also be from having actively gone and checked the working.


Similarly, "Trust - but take care in whom."


What do you recommend? Without a criterion for trusting some previous work this would make science impossible.


"uneducated" is committing to the same fallacy that you are trying to argue against. Part of the current political climate is precisely about the lack of information, arguments etc. You might call that a concern of education, but the real problem is learning, which is a far broader problem than supervised education.


The problem comes from people conflating the term 'amateur' (does not pursue a paid-for career in the chosen field) with 'ignorant' or 'unskilled'.


The Industrial Revolution came about because The Enlightenment required that every argument was judged on its merits.


There's a lot of pseudoscience out there like flat earthism, homeopathy, dowsing and what not. The belief that "the establishment" is ignorant and narrow-minded and could be proven wrong by a determined individual willing to challenge those beliefs is what enables these beliefs in the first place.

Skepticism and curiosity are healthy, driving forces of science that can be taken too far to lose touch with reality.


In addition to the earlier points, I find a lot of the reporting on the misuse of statistics in psychology misuse statistics. It's a narrative, so "psychological studies are underpowered" becomes "ARE ALL PSYCHOLOGY STUDIES FAKE????"


In fairness, you can find psychologists seriously asking that question too, going back decades.


I guess I meant in regards to studies that suggest psychology studies are underpowered (there have been quite a few in recent years), reporting often says that the studies have been disproved, which most of the time isn't statistically correct.


Not just that.

> Of course, psychologists often use their research findings to speak about how psychological processes function. But, such inferences represent a basic misinterpretation of our analyses.

> Professor and Chair of the Department of Psychology... at the American University of Paris

https://blog.oup.com/2017/08/psychology-silent-crisis/


I agree that some things in academia are overly fenced off from the influence of an educated everyman but I don't think that applies in this scenario. The article is specifically about how the student recognized a problem but also recognized that they had particular shortcomings that made tackling the problem adequately on their own impossible.


>Sure wish folks would quit sucking on the "amateur vs. the establishment" teat.

That's just a counter-balance on the enormous "sucking on the establishment's teat" going on -- especially when the establishment is such vacuous and BS as "soft sciences".


Hey, as if soft sciences in the broadest sense weren't hard. They are hard to learn, because they are dealing with hard problems, for example how to raise a child or teach a whole heterogeneous classroom with minimal effort?

It's certainly not an ideal name.


Soft not as in "easy to learn", but as in "not very solid".


Fluid dynamics? :)


>"Just as zero degrees celsius is a special number in thermodynamics," wrote Fredrickson in Positivity, "the 3-to-1 positivity ratio may well be a magic number in human psychology."

Well this statement still stands.


It's worse than that.

0C is at least special in in that important human systems of measurement are based on it. The 3-to-1 ratio is just some ill-fated psych superstition. It resembles the Celsius scale only by being a cultural construct. But by that measure it also resembles South Park and the MS-DOS kernel.

And that's ignoring the fact that 0C is actually physically significant, since it is (to a good approximation) the melting point of water over a very large range of pressures.


I don't understand, what is the special status of 0C?


The author is confused. 0C is special to us because it is the freezing point of water, but that doesn't really make it special within thermodynamics -- it's just the freezing point of some arbitrary substance at an arbitrary pressure (ATP). 0C isn't even the only freezing point of water -- if you add impurities or change the pressure, you can change the freezing point!

Maybe 0K (Kelvin) is special, since that implies zero gas entropy, but 0C doesn't have any magical properties.


Interestingly, unlike the boiling point, the melting point is not that dependent on the atmospheric pressure: for a wide range around 100kPa, it stays at 0C [1].

[1] Wikipedia: https://en.wikipedia.org/wiki/Pressure_melting_point


Well, the irony is that it isn't very special, and neither is the positivity ratio. So the "just like" statement turned out to be true after all. ( A nxor B == not A nxor not B )


I'm sure @Maybestring is being ironic.

Neither are special.


The post was a comparison not a contrast. I believe the point is that the 3-1 ratio has no special status - not that 0C does.


Neither are special


Maybe they meant Kelvin?


Probably would have been a better metaphor had they used Pi.


https://arxiv.org/pdf/1307.7006.pdf Such an amusing read!


You have to respect (fear?) a paper that Sokal and Pinker have helped with.


With Gelman in there too? Definitely fear.


Fun fact, one of the big guys in positive psychology accidentally inspired the Bush torture program, which was similarly based on scientific-sounding nonsense.

https://www.newyorker.com/science/maria-konnikova/theory-psy...


That's like saying that our understanding of respiratory anatomy inspired water boarding.

I'm glad New Yorker has improved its journalistic standards a little since then.


Typically in U.S. psychology majors, statistics is required and is taken in the Spring of the student's senior year. I've been playing with the idea of having statistics be the second course (after intro) in a psychology major so that all other courses could build on the students' quantitative understanding.


Having taught the topic for almost a decade, I know that both among undergrad and grad students, Stats is viewed as a roadblock by most. In addition, most people who teach the topic, don't want to teach it, so they do it in the easiest way possible: Compute this number, look up this table, tell me if the result is "significant" ... "There, you get an A. One less obstacle on your path to becoming the best doctor, the best journalist, the best I-banker, the best consultant, the best of everything ... To bad you don't know what this all means at all."

Edited to add: Also, I can't bring myself to forget the positive reception Sam Wang got here for absolutely meaningless statistical statements. But, again, most people see Stats as providing the tools with which to make such statements so they can be popular.


It seems to me that stats and scientific method are the core of psychology. All the domain-specific content is just common sense. This quote from Brown says something "Not many psychologists are even good at the maths and statistics you have to do as a psychologist.". Perhaps even rename psychology to "statistical study of fuzzy ideas" or something.


Statistics is not the core of psychology, it is the core of having a publishable result. Psychologists want to do just enough statistics that they can claim to have a p-value<0.05 so they can get it into a journal. They don't care if their assumption of normality is correct, or if their assumption of independence is correct, or if the question they asked on the survey is actually measuring the thing they want to measure (spoiler alert: it usually doesn't, but we have a hard time measuring the brain). So they willfully ignore the assumptions because acknowledging that they were violated would mean that they would have to learn more complicated statistical methods. The current use of stats in psychology is just an attempt to make it look like their research is rigorous, when in reality it is usually broken.

The ideas of the scientific method are fine in psychology, but since most studies use WIERD samples, don't be surprised that many of those results don't generalize to other demographics.


Statistics is barely complex math, hardly more than Algebra formulas one plug in numbers. The theory behind the formulas are the foundations of logic, so the idea that even any significant fraction of professional psychologists do not understand statistics is damning of their their profession. That is similar to accountants being bad at addition and subtraction! Astounding.


I am afraid you don't understand what makes Stats hard.


I experienced how poorly statistics is taught in American Universities. I took the entry class twice, and never felt I understood. Perhaps that is my point, I knew I needed to understand statistics to do any serious work I was interested. I ended up taking a full summer, 12 credit hour, 5 hours a day, 8 week Seminar to gain the level of understanding I knew would be required to make a mark. Statistics is not hard, nor difficult, in the least. It is taught with utter disrespect in the US University system, with a "you don't need to understand" attitude that is 100% backwards.


As an outsider, the most likely culprit seems to be american school system.


Even so, bsenftner, has shown that many people in a field that uses statistics don't understand some of the very basics in their field. Implicit in that, many other research papers in that field are likely to be just as flawed as the one discussed in this article.


I am the person featured in the Observer article (for "proof", see the account with this username on Twitter). AMA. :-)


What are your thoughts on the "rest" of Frederickson's research? You seem content to debunk only Losada's math (charitably characterized as a "brain fart"), where this article seems more willing to lambast positivity research on the whole as self-serving mumbo-jumbo.


At the time that interview was published I didn't know too much about it. Since then I have published about 5 articles critiquing various other work from the same lab. It mostly seems very weak, but I'm not sure if it's any worse than most of research in the wider field of social psychology (and peripheral subfields that use similar underpowered research methods). For example, I have also been involved in looking at the work coming out of the Cornell Food and Brand Lab (GIYF) and much of that seems to be catastrophically bad.



Can you give us a quick run-through of the maths that led to the 2.9013 number?


No, especially five years on. :-)

Seriously, there's no obvious way to do this in a "quick run-through" that is quicker than what we wrote. Look at the calculations in our article on arXiv (and if necessary, look up the 1962 Lorenz article and find the relevant equations). 2.9013 is actually an abbreviation for a recurring decimal representation of a rational number(!), which on its own ought to be a sign that something is wrong.


Thanks; the closest a quick search threw up was 22987/7923 = 2.9013000126 which confirms the bogosity - I was just interested to know how the bogosity was being justified.


It's great that this has been picked up recently, it's something I've been following along with for a while. If anyone is interested, James Heathers runs a podcast called Everything Hertz (https://twitter.com/hertzpodcast) with Dan Quintana and Nick has been a guest on it - it's well worth listening to if you're interested in science full stop.

These guys aren't researchers in my area of interest but the topics they cover are interesting and done very entertainingly!


Nice to see Sokal's name come up, he's the first person I thought of when the article outlined the obvious bullshit use of advanced maths in a psychology paper. It's amazing how far stuff like this can get within a field before someone gets round to calling it out.


Another physicist, David Deutsch, also does a good job in The Beginning of Infinity, Chapter 12 'A Physicist's History of Bad Philosophy'. I can't do it justice here but part of the problem is that we don't understand what 'happiness' is yet and therefore it's a problem how to measure it, let alone self-report it.


An amateur who did the work. He was doing a relevant degree and collaborated with people with more experience or expertise in relevant fields. That's how you provide useful critique of scientific theories.


One thing I'm confused about is why most of the criticism has come on the Positive Affect paper and not the original "The complex dynamics of high performance teams" paper, which the most of the math is drawn from.


A common topic I see is the surfeit of trained scientists, and how many of them are unable to find employment in their field (professor or other acceptably remunerative research position). Maybe many of these guys out there are simply not good enough to do science, and we should raise the bar for entry to these training programs if so many of their graduates fundamentally misunderstand the mathematics their analyses rely upon.


> "Each of them appeared to quote and promote one another, creating a virtuous circle of recommendation."

There should be a word for this - is there? It runs like a red thread in all sorts of dubious endeavors in the public sphere.


Citation cartel (https://www.frontiersin.org/articles/10.3389/fphy.2016.00049...). Basically, the equivalent of a webring (https://en.m.wikipedia.org/wiki/Webring), but stealthy and on steroids.


ah - "Citation cartel". Thanks!


In literary circles the term is "log rolling". That term seems to be less popular today to refer to public figures endorsing one another, but in literary critiques it is still used.


Thanks for the term! Yes, in literature and in other culture areas is exactly where I've seen it most often too, and it has become a pet peeve. It is conceivably more harmful in science (not to mention politics) however. Harder to spot though.


self-congratulatory ring?


Circle jerk?


It's basically a form of Cronyism, promoting each others work without regard to merit or qualifications.


I read to the end of the article hoping to find out what exactly the error in the math and/or reasoning was. Apart from some vague mentions of how the "tipping point" might be influenced by other factors, there was no detail. This article is largely content-free.


95% of published papers are bullst and you know it the moment you read the abstract. The problem is having the time and energy along with a healthy disregard for the gatekeepers on your academic path to keep pursuing this.

That's why it usually takes an outsider to publicize this stuff.


When I think about how most papers are written by a student, with credit given to their professor, and I think of what a mess is code written by developers the same age as those grad students... It’s amazing we get as much right as we do.

And I know we are getting things right because I have a phone in my pocket that has surmounted thousands upon thousands of puzzles in physics and chemistry over the last twenty years to do what it does.


I’d wager that most papers in most fields are bullshit, but it doesn’t matter because no one reads them or bothers to attempt applying them or building on them.


Didn't you mean to say 95.274% ?


Reminds me that 47.72% of all statistics are completely made up.


In my field (robotics, ML) many papers are not interesting or overstate the consequences of the result, but I wouldn't go as far as to say 95% are bullshit.


> If your ratio was greater than 2.9013 positive emotions to 1 negative emotion you were flourishing in life.

I cannot fathom how honest, intelligent people end up thinking they've "solved" for something like happiness or flourishing in a mathematically meaningful way. And to 4 decimal places!


I am just picturing someone haranguing a person for dropping them from 2.92 to 2.90 and ‘ruining everything.’


You should watch Black Mirror s3e1 "Nosedive"!!


Look,

Puppy dog, Birthday, Fun at the beach.

Are we cool now?


Is that still the default precision displayed in Excel?


By that logic, if someone has 3 happy moments and then kills themselves they are one of the happiests people who has ever lived.


My own (non-scientific) opinion is related to this. In my experience, resistance to negative emotions is a much more crucial ingredient to happiness. As I’ve been gradually maturing during my life, I think the biggest improvement is that I’m quite emotionally stable. Small and unimportant things don’t bother me much at all, which really helps when it comes to leading a happy and fulfilling life. Especially as negative experiences are supposed to have a greater impact that positive ones.


Only if you could successfully argue that anyone would kill themselves in those circumstances. I won't hold my breath.


Who wouldn't want to be the happiest person to ever live? That seems to be a happy moment in itself.


Because there's the possibility of being even happier. Just because we're happy, doesn't mean we're not greedy.


Aha! So a better happiness equation is both cumulative and a ratio.

If G is the cumulative number of good moments and B is bad moments, happiness is (rounding the happiness ratio to 3):

happiness = (G/B-3)*G = G^2/B-3G

Seems there could be a physics of happiness from this equation.


It's too bad the field of positive psychology is left to people like this, because it asks useful questions and would benefit a lot from serious research.


>"Just as zero degrees celsius is a special number in thermodynamics," wrote Fredrickson in Positivity, "the 3-to-1 positivity ratio may well be a magic number in human psychology."

I can't tell if he does or doesn't understand thermodynamics.


I'm assuming she was thinking of zero degrees kelvin.


Here's the nub of it:

> "Not many psychologists are very good at maths," says Brown. "Not many psychologists are even good at the maths and statistics you have to do as a psychologist."

It's amazing -- well not at all amazing really! -- that the most successful people in any scientific field come from an engineering and physics background.

One of my favourite quotes is from Kelvin:

> I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science, whatever the matter may be."

And ... mentioning Kelvin is a nice segue to this quote from the article:

> "Just as zero degrees celsius is a special number in thermodynamics," wrote Fredrickson in Positivity, "the 3-to-1 positivity ratio may well be a magic number in human psychology."

Almost Sokal-ish in its attempt to borrow some relevance. Zero C is not very special. Zero Kelvin is!


I've seen plenty of computer science students who struggle with math, and even among those who the number who understand statistics is embarrassingly small. Meanwhile, here on HN you have dozens of "computer engineers" arguing that the maths and theory in their own field isn't important.

I wouldn't be so quick to pull generalizations about what disciplines have better scientists.


Big part seems to be that statistics often isn't properly taught to students, which then muddle through it when writing papers.


You are conflating research with practice.

Most computer engineers are programming and can get away with not being up to speed with calculus/stats. As can most practicing psychologists (to a much greater degree).

Phd computer scientists doing research are typically not 'weak' at maths.


I have a friend who is a linguistics professor at a major public university, and she has complained to me about some of the projects she's been involved with where her fellow researchers really are not good at math.

Among other things, there's a lot of confusion about statistical significance, and how and when it's calculated. There are people out there, especially in the cognitive and social sciences, who think it's OK to run an experiment to gather some data, and then look for statistically significant trends in the data and draw whatever conclusions they can. They're p-hacking and they don't even know they're doing it; they think it's science. And the people judging their papers sometimes think the same thing. How do you fix a field like this?

This issue lies at the root of the reproducibility problem. The example someone gave about an undergrad cheering (that she didn't have to do any more math) points at one of the causes.


> It's amazing -- well not at all amazing really! -- that the most successful people in any scientific field come from an engineering and physics background.

This seems like a pretty wild claim to me- what is your basis for it?


I probably overstated it. I meant that engineering and physics people are unexpectedly* successful in other scientific disciplines.

* I know that's pretty subjective.


Is it possible that you perhaps have an engineering or physics background (or one sympathetic to the same- say mathematics or computer science) and pay more attention to others who are like yourself? Would you make similar statements regarding people from, say art backgrounds who make great contributions in engineering, physics and other sciences-Morse, Hedy Lamarr, Santiago Cajal to name a few? I think it's pretty common for clever people who do well in one field and decide to switch to do well in another, no matter what the two fields are.


I agree, this is pretty wild, but I also agree with the sentiment. Having studied physics I might be biased, so help me out here, please. Name one scientist from the 20th century more popular than Einstein, groups and abstract ideas are OK, too.

PS: "most successful" is ambiguous. Success is not really comparable. You can have many small successes, revolutions in fields that today feel like common knowledge and not scientific at all, and many nameless contributors. Medicin, chemistry, and many interdisciplinary fields.

Fixating on a single field only to argue in the end that all other fields are basically just subfields of ... metamatics, is funny, but missing the point. There is a large perceived disconnect between the natural sciences and "others". That's the real problem OP was pointing out, I guess.


It is certainly aviability heruistic[1]. You need no external help to see it, just try to remember 5 physicists and 5 psychologists and compare efforts it takes.

Someone to compare with Einstein? Freud. This guy is no less known to public, and influenced society maybe even more, by breaking taboos around sexual desires. Moreover, he shaped all the modern psychology.

[1] https://en.wikipedia.org/wiki/Availability_heuristic


But what was Freud right about?

Another measure of the success of a scientist is how well their theories stand the test of time, and by that standard, mathematical rigor probably contributes a lot.


> But what was Freud right about?

What was Freud wrong about? His theory is a model of human mind, which has its application even today and suits it well. It is like Newtonian mechanics and gravity: did Newton was wrong? No, he developed a model of reality that works even today.

There are a lot of other theories which replicate psychoanalysis in many aspects. Even humanistic psychology highly influenced by psychoanalysis despite of (or in spite of) negativism of Carl Rogers who didn't liked idea of explaining all of the mind work by primitive impulses. Rogers just made new model of mind by changing words and turning psychoanalysis upside down: mind is driven not by primitive impulses but by the need of self-actualization. The same eggs in a side view.

> how well their theories stand the test of time

What it means to stand the test of time? It is social sciences, not the math or physics, the landscape here is constantly shifting, nothing is true and nothing is false. The only way I see to measure the success of scientist is to measure its influence on others. If so, then Freud is the total winner among psychologists. No one even come close to him.

There are others -- Skinner, for example. But even Skinner's behavioral approach moves towards Freud ideas: now it is cognitive-behavioral approach, with hidden cognitive processes and conflicts between them. What the difference between coping strategies and defence mechanisms?

I'm not trying to say, that psychoanalysis is the best, I do not like it really. As for me it oversimplifies things. But I acknowledge influence of psychoanalysis on modern views.


You haven't answered what Freud was right about. I do not deny that Freud and psychoanalysis were influential - so was Star Wars - or even that psychoanalysis can help people in specific circumstances - so can yoga - I deny that he was a scientist.

> What was Freud wrong about?

To cite just a few examples: women (penis envy, hysteria, female orgasms); psychosexual stages, including the idea that homosexuality is anal fixation; the id, ego, and superego; the oedipus complex; schizophrenia.

> nothing is true and nothing is false

Then it's not science.


> Star Wars

Few people in Asia have ever heard of it. It's not as influential or popular on a global scale as one might think.


> You haven't answered what Freud was right about.

I don't know how to answer this. For example, he proposed two structures of mind topographical and dynamic (I'm not sure how they named in English, and I'm too lazy to search), and proposed how that layers interacts. His ideas of psychological defences, inner conflicts... Is it counts for "Freud was right", or it is too obvious now and does not look as important ideas? All the views on a mind working are highly influenced by Freud, sometimes it is hard to find his ideas in our views, because we are too accustomed for them and therefore we are unable to see them.

> To cite just a few examples: women (penis envy, hysteria, female orgasms);

Yes, and no. Histeria and similar conditions even for now are treated with regard to psychoanalitic views -- unconscious conflicts based on repression or some other psychological defence. Psychosomatic -- I attended for classes in psychosomatics and sow patients with psychosomatic conditions. The behavioral approach does not do well, because it ignores the roots of conflit which lead to strange behaviour, and one cannot cure anorexia or hysteria relying on drugs or behavioural training. Behavioural approach can change sympthoms, for example, to make a bulimia from anorexia. Or help patient to regain control of paralized hand, but then patient would start having seizures. If conflict had not resolved, than it manifest itself somehow. Patient needs to go through the breaking of psychological defenses to understanding of conflict and to resolving that conflict. The only signigicant difference from psychoanalysis is that conflict does not nessessarily comes from psychological trauma experienced at age of 6 month.

> psychosexual stages, including the idea that homosexuality is anal fixation

There are a lot of authors who developed that ideas and the stages of developmet remained the same in general, explanations was changed.

> the id, ego, and superego;

Whats wrong with them?

> the oedipus complex;

Yes, I also do not like it. But a developing sexuality of a child and a sexuality of a parent do interact somehow. The oedipus complex seems not the best way to describe this interaction, but Freud was the first researcher in the field. If we can call him a researcher -- his empirical methods was controversial indeed.

> schizophrenia

And even now no one knows what shizipheria is. Seems that diagnosis "schizophrenia" is a way to say "we don't know whats wrong with him". I agree, Freud was wrong about schizophrenia, just trying to point that modern authors also do not know.

> Then it's not science.

I think I understand your point, because I moved from tech and math into psychology, and years ago I also could say something like this -- "it is not a science". But now I think that this phrase means "is not the physics and it is not the math". Yes, social sciences are different. They are continuously failing to make formal models, written with pretty greek letters in the math language. Lack of formalization leads to a lot of problems, but I see no way to formalize the field. And I know that there are a lot of people who are much stronger than me with math and they do not see the way to create a formal models either. There are formal models for local phenomena, but very limited ones.

The lack of formalization comes from complexity of the object of research. Should we claim that science that failed to be a formal and strict because of complexity is not a science? If so, than what we should do next? To stop any attempts to make a science in this area?

Human mind is much more complex object to deal with in research than quarks. At least there are only six flavours of quarks, while all humans are different. All the social models are models which works only conditionally. In many situations no one even know what that conditions are. For any given model we can find conditions when that model would give us correct predictions. And for any given model we can find conditions under which model would fail. It is means that everything is true and everything is false. Boolean logic fails and the law of excluded middle does not works.

Social sciences are dealing with problems for which science has no ready-made tools. The reproducibility crisis is a social sciences invention, it is impossible in physics and is inevitable in social science. Now social sciences are searching for tools to fight it. Physics does not need such a complex tools, because it digs deeper, not forward. Physicists laugh at philosophy now, because they need no philosophy any more. Philosophy now all about social science problems.

So, maybe social sciences are not sciences. But at least they are meta-sciences, they are trying to build science which would be able to deal with social problems in scientific ways. And we have nothing to replace social sciences in their area.


> Lack of formalization leads to a lot of problems, but I see no way to formalize the field

It seems that economics and sociology do play a role there. Both are heavy with statistics and game-theory. While Psychology is only looking at individual. Humans are more complex than fundamental particles and molecules. It's not chemistry or biology either. But Neuro-Science is most promising in this regard. "Nerven-Arzt" is a dated German term for Psychotherapist, after all.

On the other hand, what would a psychologist say to the urge to formalize everything? :)


> mathematical rigor

that's very stiff.


Thank you!

There is a thing to say about the beauty of maths as a universal language. In that spirit, I see Freud closer to philosophy, in the sense that philosophy is highly divided and regional.

Was freud really that influential? The "freudian slip" is certainly a very useful tool for analysis. But his "psycho analysis" is outdated, or at least overcome. And sexual freedom is progressing far slower.

Sexuality is a baseline of human development. You made a very good point.


His psychoanalysis maybe outdated but even now you can find orthodox psychoanalists (not just modernized version). Also there are a lot of schools that derived from orthodox psychoanalysis and moved away for some distance small or large. I believe that any psychoterapist know at least basics of psychoanalysis, can explain about id/ego/super-ego, psychological defence mechanisms, psychological resistance, transference and countertransference. There are bunch of theories of human development derived from Freud's theory psychosexual development. Even if psychoanalysis is dead (which is not true) it exists in derived works.


I do not generally measure the effective "success" of a scientist by their brand name recognition.


Science is a popularity contest now?


Freud? Marxism?


> "Not many psychologists are very good at maths,"

Most social scientists are not that good at maths. It used to be mostly harmless before they started dabbling in experimental science involving statistics.

A bigger problem is Economists; many of whom are also bad at maths, but litter their articles with formulas anyway, to add some pseudo-intellectual rigour.


Economist here. I completely agree


While Kelvin claimed that "when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind", it is interesting to note that he also maintained that "I believe that the more thoroughly science is studied, the further does it take us from anything comparable to atheism."

https://books.google.com/books?id=kdm9nmAkKvIC&pg=PA1103&lpg...


> "Not many psychologists are very good at maths," says Brown. "Not many psychologists are even good at the maths and statistics you have to do as a psychologist."

It's worse: IME in grad school, the field is politicized beyond reason. It was a very depressing realization.


Are you referring to the field's own internal politics (e.g. preference to psychological doctrine or against, e.g. BF Skinner), or are you referencing American government politics? (e.g. like Democrat or Republican)


Internal politics - both within the field and within the department.

- The main focus of my advisor was a life-or-death struggle with a researcher at distant university over the number of angels that dance on the head of a pin - utterly inconsequential, self-justifying research and counter-research that ultimately served nothing more than to keep the 'publish-or-perish' wolf from the door. Just enough math was involved to keep it from being sociology.

- A grad student with an engineering background was dismissed for discretely pointing out that the measurement tools for one professor's lab's flagship experiment series were an order of magnitude too inaccurate for the measurements they were being used to perform. This was commonplace - another was warned for pointing out (privately, not in public) that requiring us to pass all our submitted papers through TurnitIn was also signing over our legal rights to anything we submitted as part of our coursework (which could potentially cause publication problems)


I think you are giving the fields of engineering and physics too much credit here. Not the least because you are assuming that engineers and scientists are by default good at math.

Math is hard. Pretty much period. Often, you can trick yourself into forgetting how hard it is. More, you can assume that mastery of something that takes high math to describe, somehow indicates mastery of the high math. Consider juggling as an easy example, can be very complicated to explain what is happening, yet I don't think it would be wise to assume that all jugglers are good at higher math. Even if the ones you find are.


Dr Feynman would agree:

    The principle is that you must not fool yourself, and you are the easiest person to fool.


Math and physics can describe juggling, but that description is not what helped discover juggling - holding a few balls in your hand suffices for that. Engineering and physics on the other hand to be math derived fields. In other words, the processes involved are sufficiently complex to be beyond the scope of even a rather clever human's intuition. Of course you might argue of success by mimicry, yet mimicry becomes exponentially more difficult as the complexity of a problem increases, to the point that I think we can all but disregard it for meaningful applications of engineering and most natural sciences.


You'd be surprised how much math can actually help get into advanced juggling. Similar to origami. Sure, basics don't benefit from math. Advanced practices, though, do.

Regardless of that, the point of the example was precisely the point you seem to be using against me. Most practices in many fields, including much of physics and engineering, don't actually need advanced math. They are helped considerably by it, though.


Yes Mathematics is hard.

I don't think being good at juggling makes you any better at describing the dynamics of juggling, nor the physiological characteristics that enable it.

Regarding Feynman (comment below), yes it is very important not to fool yourself, but knowing mathematics well makes it easier not to fool yourself.


But does it? Or is it just that the people you know that don't fool themselves happen to be good at math? :)

The juggling example was supposed to be an obvious example of people doing something that takes complicated math to analyze. I'm asserting it is not much different to many engineering jobs. Nor do I think that most physicists have a terribly strong intuition for math.

I confess to not having numbers to back this. :(


How does Kelvin quantitatively show that only quantitative claims count as true knowledge?


Kelvin is making a philosophical claim about scientific knowledge. There is no contradiction here.


You're overstating Kelvin's claim. He doesn't say it is the only way to true knowledge.


> "Not many psychologists are very good at maths"

Proof?


As a general response, I would point out that the entry requirements to study Psychology at University are lower than those of more 'hard' mathematical subjects, such as maths and physics (at least in the UK). During their period of study students also study the bare minimum number of mathematical modules (mostly statistics), therefore, you could argue they are not fully trained in 'maths'.

A completely anecdotal experience: During my undergraduate (MPhys) I lived in halls with about 20 psychologists. I remember one night during the second year (of 4) they all had a huge celebration. The reason for the celebration was that they had had their last ever maths exam of the undergraduate. I remember one girl proclaimed "Thank god, I've never have to do any maths ever again!". She was met with a roar of support... I couldn't help but cringe thinking about the necessity of statistics in their future research.


As someone who studied Psychology and CS, and is somewhat good at math, it's not necessarily a bad thing that psychology students don't take a lot of math.

I'd say maybe 10% of students who take psychology actually go on to do anything with research. Most of my friends in psychology are now counsellors, social workers, occupational therapists etc., and are good at their jobs, and this would not change if they had to take multivariate calculus.

At the same time, I do think psychological research would benefit if the people performing that research were better trained in math (actually when I say math, I specifically mean statistics and linear algebra).

I think the problem is that psychology is so broad that it can't possibly do a good job at catering to all these different concerns. Even at the graduate level, in order to become a clinical psychologist, you have to do a lot of research. The scientist-clinician model sounds good, but in practice I found that a lot of people who just didn't care about research were doing research. I found the same thing with medical students who were doing research to pad their resumes.

I also think that you can't just take a physicist or mathematician and plop them in psychology and start fixing everything. The problems often require a ton of theory, are really expensive to test, and the data quality is often terrible (because of human error, measurement error etc). Relevant XKCD: https://xkcd.com/1831/


>I would point out that the entry requirements to study Psychology at University are lower than those of more 'hard' mathematical subjects, such as maths and physics (at least in the UK)... During their period of study students also study the bare minimum number of mathematical modules (mostly statistics)

Because the math is easier, and all you need is statistics. As the article states, the problem is, it seems even that lower bar isn't being met.


> Almost Sokal-ish in its attempt to borrow some relevance. Zero C is not very special. Zero Kelvin is!

I am sorry if I give the impression of nitpicking but I live in a Celsius world and zero c is special because it means water freezes. (and snow instead of rain)


Zero Kelvin is still a different league of special scientifically speaking.


It's special in our human world, but not in thermodynamics / physics. Water is just a molecule like all the others, it's not special.

Whereas 0K is the absolute minimum temperature, so that's a really special value!


I'd say water freezing is still interesting thermodynamics/physics :). (But I agree 0K is another kind of mark)


I see where your coming from here, and by all means 0 °C is more significant than say 0 °F. But water freezing at 0 °C is far less "special" than everything freezing at 0 K


Sheesh, people... is the downvote really necessary ?


> Zero C is not very special. Zero Kelvin is!

I agree that Fredrickson is grasping for relevance. However, 0 °C is special to water-based lifeforms.

0 °C is defined as 273.15 K, or the temperature at which the three phases of water are in a thermodynamic equilibrium — at a pressure of 1 atm, or 1 standard atmosphere.

The degree Celsius is defined as equal to the kelvin, so the Celsius scale is a shifted version of the Kelvin scale.

https://www.bipm.org/en/publications/si-brochure/kelvin.html

https://www.bipm.org/jsp/en/ViewCGPMResolution.jsp?CGPM=10&R...


Isn't 0.01 °C the triple point of water (the temperature at which the three phases of water are in a thermodynamic equilibrium)?

The triple point is a better standard and is used now, but it's not exactly 0 °C because that was defined by calling the melting point of water 0 °C.


> 0 °C is special to water-based lifeforms.

Only those who happen to live at exactly 1 atmosphere of pressure.

And ordinary (unpurified) water doesn't freeze at exactly 0 anyway.


Actually, to off-topic nitpick, the triple point of pure water is at 0.01°C (273.16K), not 0°C


I've always thought that English / many other humanities departments learned to avoid much of the controversy that plagued Psychology departments by simply avoiding any data altogether.

If Fredericks and Losada had written nearly the same thing, using the same evidence but without the fancy maths, it would no doubt be accepted and lauded within English departments. Any attempts to fight it could be battled in the same way the humanities departments responded to Sokal in the 90's: by derriding him as "a pedant, a literalist and a cultural imperialist".


I don't know if that's true. But it reminds me of a joke:

The university is, as always, looking for ways to cut the budget. The vice chancellor visits the head of the mathematics department and asks him what resources he might be able to do without. "Resources? The only things we mathematicians use are a pad of paper, a pen, and a waste paper basket!", the head of mathematics replies. "Perhaps you could learn from the philosophers", says the vice chancellor; "they don't use the waste paper basket.".


I do not think your argument fair. In fact I find it disappointing. Some things lend themselves to collecting numbers and figures and measurements, and other things quite simply do not. Philosophy, history, literature, and human languages are what first come to my mind as important fields of study that have significant impact on our outlook on the universe, but we can't scientifically measure them. Indeed it is philosophical thought and ideas that underpin our modernity. Moreover, much of our collective advancement is in significant peril because 'the arts' have been devalued so much. (Although William H. Miller just donated a significant sum to the John Hopkins philosophy department since he felt his degree in philosophy made him so successful.)

What you go on to speak of in your second paragraph is the art of rhetoric, and ironically you have unwittingly engaged in the very sort of ad hominem attack which you are complaining that you think that liberal arts scholars would use to defend their position.

An ad hominem attack is a classic logical fallacy and that is something that I suggest you research.


Unfortunately, in order to make certain claims, you have to have metrics. Just because something is hard to measure doesn't mean you can forego it.


How do you measure the impact of John Locke, David Hume and Adam Smith?


> Philosophy, history, literature, and human languages

My claim was that some fields avoid data altogether so that they can't really be "disproved", only disagreed with. But in these cases, there are rigorous approaches that rely on data: e.g. analytic philosophy, rigorous history, linguistics, etc. and those approaches generally have a large impact.

In linguistics, Chomsky made such a huge impact on human languages because he had developed a rigorous grammar and was able to find counterexamples to many of the existing hypotheses. Chomsky had made testable hypotheses, which could only be backed up or disproven by rigorously studying human languages. And the most recent counterexamples to his grammar (e.g. that some languages lack subordinate clauses) augment rather than disprove it.

Similarly, advances in philosophy largely came from rigorous approaches -- Russell's attempts to formalize logic (and somewhat subsequent failure) resulted in Wittgenstein's formalization of the relationship to language, which had a huge impact on subsequent philosophy.


If you read Quine and Kuhn, you will find that natural science results also cannot really be disproved, only disagreed with.

It is a bit funny to see anecdotic evidence given to support the claim that empiric research is superior. Shouldn't you present data instead of anecdotes?


[flagged]


This is harsh, but true.


When are we as a society going to clearly point out that psychology and related disciplines are not science, and that young people shouldn’t confuse them with science?


Probably never given how broad the field of psychology is. The demarcation line probably runs somewhere through the field, but to separate the two sides between science and not science is an open problem. Hell, it cosmology and evolution also suffer from similar problems, but those fields are far less controversial.


Yes. As a psych major in college, I was astonished to find that the same major included, say, quantifying the changes in neurons in rat brains after rats were exposed to specific stimuli, and empirically unmoored speculation about supposed universal desires to have sex with one's mother.


I deserve the downvotes: of couse you can measure whatever you like and conduct a valid analysis of the data and that's science.

But given the extraordinary prominence of quackery in 20th century psychology and psychiatric medicine, including the grotesque over-application of psychiatric drugs to young people suffering from feelings that can and should be treated more traditionally, I do think it would be best to step back and completely excise the fields as they exist from what we call science, and start afresh.




Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: