Science Fictions: How Fraud, Bias, Incompetence, and Hype Undermine the Search for Truth
- Downloads：9422
- Type：Epub+TxT+PDF+Mobi
- Create Date：2021-04-21 12:53:41
- Update Date：2025-09-06
- Status：finish
- Author：Stuart Ritchie
- ISBN：1250222699
- Environment：PC/Android/iPhone/iPad/Kindle
- Download
Summary
Hypeology will expose the bias， hype， incompetence and fraud that plague the peer-reviewed world where many of the most seductive and striking scientific studies originate， and will take a Freakonomics-style look at the implications of this crisis for us all。
A whole industry of books and TED-talks has been built on the findings of psychological studies， presenting them as a toolkit we can apply to our lives。 That industry proved resilient in the face of the ‘Replication Crisis’ of 2011 that revealed hundreds of famous experiments to be unrepeatable， dramatically undermining their conclusions。 But as Dr Stuart Ritchie will explain， far from being solved that crisis has since spread to other disciplines。
In one large-scale organised replication effort in 2015， in which psychologists repeated 100 experiments from top academic journals， only 39 of those papers’ results were successfully replicated。 Another recent analysis found that almost half of 30，000 scientific papers published in peer-reviewed journals contained numerical errors; 15% of them contained errors so severe that they entirely flipped the results described in the paper。 A 2016 survey of scientists across all fields found that 52% believed science is facing a “significant crisis”。
Hypeology will explain how bad practice and dodgy results have become endemic in contemporary science， and it will provide a devastating and gripping take-down of the over-simplified way many of us – including scientists – present and interpret even those results that are reliable。 It will be a celebration of everything that is great about science， a manifesto for a better and more rigorous way of thinking about scientific data and its importance for society， and it will provide the reader with a toolkit for spotting bad science for themselves in order to save science from itself。
A whole industry of books and TED-talks has been built on the findings of psychological studies， presenting them as a toolkit we can apply to our lives。 That industry proved resilient in the face of the ‘Replication Crisis’ of 2011 that revealed hundreds of famous experiments to be unrepeatable， dramatically undermining their conclusions。 But as Dr Stuart Ritchie will explain， far from being solved that crisis has since spread to other disciplines。
In one large-scale organised replication effort in 2015， in which psychologists repeated 100 experiments from top academic journals， only 39 of those papers’ results were successfully replicated。 Another recent analysis found that almost half of 30，000 scientific papers published in peer-reviewed journals contained numerical errors; 15% of them contained errors so severe that they entirely flipped the results described in the paper。 A 2016 survey of scientists across all fields found that 52% believed science is facing a “significant crisis”。
Hypeology will explain how bad practice and dodgy results have become endemic in contemporary science， and it will provide a devastating and gripping take-down of the over-simplified way many of us – including scientists – present and interpret even those results that are reliable。 It will be a celebration of everything that is great about science， a manifesto for a better and more rigorous way of thinking about scientific data and its importance for society， and it will provide the reader with a toolkit for spotting bad science for themselves in order to save science from itself。
Reviews
Don
，
bias bias bias of not science
Alex
，
the audience of this book is clearly lay people with no idea what's going on in science and you know what， fine， this is probably decent for that audience， in the same way that middle school teaches you a lot things that aren't really correct but whatever， you have to start somewhere。 this is a sloppy presentation of the mainline narrative that emerged from the replication crisis that is uncritical and myopic。 the presentation of statistics is particularly painful。 it is hard to take seriously a the audience of this book is clearly lay people with no idea what's going on in science and you know what， fine， this is probably decent for that audience， in the same way that middle school teaches you a lot things that aren't really correct but whatever， you have to start somewhere。 this is a sloppy presentation of the mainline narrative that emerged from the replication crisis that is uncritical and myopic。 the presentation of statistics is particularly painful。 it is hard to take seriously an account that presents psychology's problems as if they are universal or new; that sees science as producing results that either correct or incorrect， rather than subject to uncertainty; that centers p-hacking and replicability as the fundamental problems; that offers no analysis that hasn't already been rehashed umpteen times; that fails to cover the vast， exciting and recent meta-scientific literature （from within psychology itself！）; that。。。 the list goes on。 this book irritated the shit out of me 。。。more
Jenine Kinne
，
Interesting look at the threats to science including the threats that create the doubt that so many people express toward science， which has been especially problematic during a pandemic。
Jackson Enright
，
Excellent book
Nate
，
Ritchie lays out the challenges that science currently faces surrounding research， funding， publishing， and the public's views on the reliability and credibility of science。 The challenges lay in the incentives and systems that we have created to support， fund， hire， and promote scientists themselves as well as the mechanisms for supporting their research。 The biases of the scientists， the popularity of hyping "breakthrough" research， the funding of "breakthrough" scientists at a higher rate， an Ritchie lays out the challenges that science currently faces surrounding research， funding， publishing， and the public's views on the reliability and credibility of science。 The challenges lay in the incentives and systems that we have created to support， fund， hire， and promote scientists themselves as well as the mechanisms for supporting their research。 The biases of the scientists， the popularity of hyping "breakthrough" research， the funding of "breakthrough" scientists at a higher rate， and the need for scientific journals to be relevant all play into creating challenges within science that limits scientific progress， wastes private and public funding， and sends the wrong messages to academia， students， and the public about the significance and capability of science。 I found Ritchie's book enlightening， if not a little disappointing， for pointing out the blatant problems as they exist。 Still I have hope where there are individuals and groups of scientists pushing for reform and improvements。 I think this is an excellent read for those who want to understand the significance or insignificance of science headlines， scientific studies， and all of the things to look for when researching how relevant a particular study or topic is。 。。。more
Marc Alexander
，
The author refers to the topic of the book as "meta-science"， I'm not a fan of the term but that's neither here nor there， and the book is excellent。Many people have heard rumblings about the 'replication crisis' that occurred sometime in 2016 in psychology。 Long story short， it turns out that a lot of （often popular， like 'power posing'） research in psychology wasn't replicable， which just means that when the studies are rerun by new scientists using the same methods， experiments， etc。， the res The author refers to the topic of the book as "meta-science"， I'm not a fan of the term but that's neither here nor there， and the book is excellent。Many people have heard rumblings about the 'replication crisis' that occurred sometime in 2016 in psychology。 Long story short， it turns out that a lot of （often popular， like 'power posing'） research in psychology wasn't replicable， which just means that when the studies are rerun by new scientists using the same methods， experiments， etc。， the results are neither significant nor strong， i。e。， the original research likely didn't find a real effect！ Fast forward five years， and the same issues have plagued a variety of other empirical fields。 Sound the alarm！ Ritchie systematically lays out the causes of the bad research。 It turns out， they're standard issues that creep into almost every area of human life/conduct。 Money， career success and status are all prominent players （and they're all intertwined）。 My general rule: when certain metrics are tied to money， career success and status， be skeptical of how those metrics are measured and how they can be gamed。 There are the issues of made up data， bias （in the form of running a variety of statistical analyses on data sets in order to find a statically significant result， which you then post-hoc make into the research focus on the paper being written） both conscious and unconscious， incompetence and hype （Familiar with 'growth mindset'？ It's effect is very small， and it only accounts for 1% of the variation in grades of students' when studied）。 If you're interested in the topic of meta-science， and how it can go poorly， then I strongly recommend this book。 It's easy to be a dogmatic science evangelist， who is uncritical of science。 This book reaffirms the importance of science， but cautions that it can go poorly in a bunch of ways for a few different reasons。 The most important changes that can be made so as to produce better science are the changes around the incentives of publishing flashy and significant results。I would've enjoyed a bit more of a dive into the nature and history of 'statistical significance' and why it's set at 95% or greater， and whether there are other means of determining when a result is really likely due to the hypothesized cause。 That might've made the book a bit more mathy， and therefore a bit intimidating for some readers。 。。。more
Simon Grimm
，
Givea a good overview of dysfunctional aspects of the scientific system。 Especially the latter parts of the book might be known to people who are familiar with p-hacking or wrong incentives within science。 At times it was a bit too alarmist and negative to my tastes。 But I might not be familiar enough with work inside academia to see its dark underbelly。 This article， by Saloni Dattani & Nathaniel Bechhofer gives good suggestions on how to improve the scientific system substantially: https://wor Givea a good overview of dysfunctional aspects of the scientific system。 Especially the latter parts of the book might be known to people who are familiar with p-hacking or wrong incentives within science。 At times it was a bit too alarmist and negative to my tastes。 But I might not be familiar enough with work inside academia to see its dark underbelly。 This article， by Saloni Dattani & Nathaniel Bechhofer gives good suggestions on how to improve the scientific system substantially: https://worksinprogress。co/issue/the-。。。 。。。more
Max
，
Hello， I'm Leonard Nimoy。 The following tale of alien encounters is true。 By true I mean false。 It's all lies。 But they're entertaining lies and in the end， isn't that the truth？The answer is no。 ― Leonard Nimoy （as guest in The Simpsons）I listened to the audiobook half a year ago and why exactly I thought this quote was relevant unfortunately fell victim to my memory half-life。 The book covers everything I'd want it to cover， I think。 I think I updated on the amount of statistical errors in pa Hello， I'm Leonard Nimoy。 The following tale of alien encounters is true。 By true I mean false。 It's all lies。 But they're entertaining lies and in the end， isn't that the truth？The answer is no。 ― Leonard Nimoy （as guest in The Simpsons）I listened to the audiobook half a year ago and why exactly I thought this quote was relevant unfortunately fell victim to my memory half-life。 The book covers everything I'd want it to cover， I think。 I think I updated on the amount of statistical errors in papers， which maybe should surprise me less given my first hand experience on the utter inadequacy of statistical education at university。 For example: Nuijten and her colleagues fed statcheck over thirthy thousand papers: a gigantic sample of studies published in eight major psychology research journals between 1985 and 2013。 [。。。] 13 per cent had a Reinhart-Rogoff-style serious mistake that might have completely changed the interpretation of their results （for example flipping a statistically significant p-value into a non-significant one， or vice versa） （Hm， though a frequent criticism of research practice I read is that we exactly *shouldn't* interpret the 5% significance threshold as something that completely changes interpretations。）Besides that one， I think I still fail to update on an emotional level on the amount of outright fraud in science， e。g。 Elisabeth Bik's findings that roughly one in twenty western blot studies included "problematic pictures"。 There should be something like a ceremony for becoming researchers where they vow on their life， soul and honor that they will never fabricate data。 。。。more
Willem
，
Essential reading for anyone involved in science， or even anyone who reads popular science sometimes
Jb
，
For anyone who ever read an article quoting a scientific study or a self-help book based on research。 Eye opening。 Some of the most famous studies are not reproducible， and it’s a systemic issue。 Love the opening from the author: seems like a combination of advocacy and technology could bring science back on track。
Professor Chris Lloyd
，
I wish there was an option to give 6 for this book。 I do not usually indulge in hyperbole but THIS IS THE BEST NON-FICTION BOOK I HAVE READ THIS CENTURY。 If you are an academic， if you are a consumer of science， if you are someone who cares about the scientific project of humanity then DO NOT MISS THIS BOOK。I made 102 notes in this book and there was hardly a critical one。 I simply could not find anything to criticise。 The reference list is meticulous and I have marked a dozen to read and extend I wish there was an option to give 6 for this book。 I do not usually indulge in hyperbole but THIS IS THE BEST NON-FICTION BOOK I HAVE READ THIS CENTURY。 If you are an academic， if you are a consumer of science， if you are someone who cares about the scientific project of humanity then DO NOT MISS THIS BOOK。I made 102 notes in this book and there was hardly a critical one。 I simply could not find anything to criticise。 The reference list is meticulous and I have marked a dozen to read and extend my knowledge。Not only is it well organised and well researched， the prose is first class and there are some terrific metaphors and occasional chuckles。 The topic is serious but the treatment is not sanctimonious。 It is forensic and well argued solutions are offered。 。。。more
Nick Bennett
，
This is an excellent reality check for everyone interested in， or working in science。 Stuart both highlights the challenges in accurately understanding scientific papers， but also brings us hope that things can get better。 Our future is science based， and if every scientist， journalist and PR professional reads this it will be a brighter future。
sssssssaten
，
Light & informal overview of the replication crisis。 Discusses several famous fraud cases with analyses of the prevalence of statistical bias （p-hacking &c）， hype （from both researchers & media）， & negligence within the scientific community at large。 The publication industry is also scrutinized as a sensationalist institution which systematically favors positive results relative to nulls & replications， perversely incentivizing careerist h-index gamification over quality research。 Ritchie offers Light & informal overview of the replication crisis。 Discusses several famous fraud cases with analyses of the prevalence of statistical bias （p-hacking &c）， hype （from both researchers & media）， & negligence within the scientific community at large。 The publication industry is also scrutinized as a sensationalist institution which systematically favors positive results relative to nulls & replications， perversely incentivizing careerist h-index gamification over quality research。 Ritchie offers some broad suggestions promoting transparency， pre-registration & other open-sourced methods to reduce the amount of unreliable science being published。 There is also an appendix offering tips to the general reader on how to interpret a scientific paper。 To any specialist， this will be useless。 。。。more
Andy Loomis
，
Years ago I had a professor who assigned E。O。 Wilson and insisted that interdisciplinary programs such as Gender Studies were inferior because their scholars were less rigorous and less devoted to truth than scholars in the "hard" sciences。 I wish I could go back in time and hand him this book。 Years ago I had a professor who assigned E。O。 Wilson and insisted that interdisciplinary programs such as Gender Studies were inferior because their scholars were less rigorous and less devoted to truth than scholars in the "hard" sciences。 I wish I could go back in time and hand him this book。 。。。more
Steven
，
Depressing but well arguedI think this book， unlike most pop science books， wasn't afraid to really go through more detailed examples。 This helped provide evidence for their depressing claim: science is broken in a fundamental way， due to bias， fraud， poor incentives among others。I really enjoyed that it came up with answers for how to fix these， but found the lack of actions / reasonable path towards fixing them compounded my sadness at science being broken。 Depressing but well arguedI think this book， unlike most pop science books， wasn't afraid to really go through more detailed examples。 This helped provide evidence for their depressing claim: science is broken in a fundamental way， due to bias， fraud， poor incentives among others。I really enjoyed that it came up with answers for how to fix these， but found the lack of actions / reasonable path towards fixing them compounded my sadness at science being broken。 。。。more
Usman Khaliq
，
review: http://usmankhaliq。com/books/science_。。。 review: http://usmankhaliq。com/books/science_。。。 。。。more
10Elton2020
，
The book makes its main points based on cases of massive frauds and invites the uninitiated reader to generalize to scientists as a whole。 It's a cynical account of how science is done， but it lays out problems with the system in an understandable way。 The proposals on how to improve science in general， the publishing process， the hiring processes， and the whole incentive structure， are given much too little space。 They feel like an afterthought， which is very unfortunate。 The book makes its main points based on cases of massive frauds and invites the uninitiated reader to generalize to scientists as a whole。 It's a cynical account of how science is done， but it lays out problems with the system in an understandable way。 The proposals on how to improve science in general， the publishing process， the hiring processes， and the whole incentive structure， are given much too little space。 They feel like an afterthought， which is very unfortunate。 。。。more
Heather Browning
，
This is a somewhat scary look at the current problems in the practice and publication of science， leading to frequently false or misleading results。 It's easy sometimes to reify the 'scientific method' and forget that it's an activity performed by normal， flawed human beings， operating within the restrictions of all kinds of social and institutional structures， and it's only by identifying the issues and finding methods to combat them that we can get to where we want to be。 It's not coming from This is a somewhat scary look at the current problems in the practice and publication of science， leading to frequently false or misleading results。 It's easy sometimes to reify the 'scientific method' and forget that it's an activity performed by normal， flawed human beings， operating within the restrictions of all kinds of social and institutional structures， and it's only by identifying the issues and finding methods to combat them that we can get to where we want to be。 It's not coming from an anti-science position， but rather a strongly pro-science stance that recognises how important science and scientists are and what can be done if these problems are addressed。 。。。more
Joseph McKnight
，
How real are all of those scientific break throughs？ Albeit it is worrisome to think that all of the science data could be called into question， I think it would be worse to close our eyes to the thoughts brought up in this book。 After reading this I wonder how broad these issues are in science。 Although the author gives ideas on how to fix this issue， I am not sure this is an easy fix。 Especially when you have studies that take decades to complete。 For a long time I have felt that scientists co How real are all of those scientific break throughs？ Albeit it is worrisome to think that all of the science data could be called into question， I think it would be worse to close our eyes to the thoughts brought up in this book。 After reading this I wonder how broad these issues are in science。 Although the author gives ideas on how to fix this issue， I am not sure this is an easy fix。 Especially when you have studies that take decades to complete。 For a long time I have felt that scientists could be less that honest when， after an intensive study， a theory is disproven。 This seems to follow that same thought process and is sad to think how deep the corruption may be。 Give it a listen and decide for yourself if the author might be on to something。 Joseph McKnighthttp://www。josephmcknight。com 。。。more
Kadir
，
It was a great source for its explanation of the current state of science and how it was improved in the recent years and can be improved in the future for a young scientist。
Nichola Raihani
，
I had been meaning to read this book for a long time and it did not disappoint。 Extraordinarily well written account of the ways that science can go wrong - as well as how we can fix it。 One of the very few non-fiction books that manages to be both informative and humorous。 I loved it。
Jesse Field
，
This book certainly reminds me of the work of Nassin Nicholas Taleb as injunctions toward skepticism。 Stuart Ritchie is sort of the Taleb of science。 Both even make mention of Daniel Kahneman’s work， though Ritchie uses it as illustration of problems in science （‘priming’ effects were found not to be defended as well as Kahnman found in his original work） where Taleb makes extensive use of the survivorship bias concept。 Ritchie is a better writer than Taleb， I find， giving us clear accounts and This book certainly reminds me of the work of Nassin Nicholas Taleb as injunctions toward skepticism。 Stuart Ritchie is sort of the Taleb of science。 Both even make mention of Daniel Kahneman’s work， though Ritchie uses it as illustration of problems in science （‘priming’ effects were found not to be defended as well as Kahnman found in his original work） where Taleb makes extensive use of the survivorship bias concept。 Ritchie is a better writer than Taleb， I find， giving us clear accounts and arguments where Taleb relies too much on fast-and-loose anecdote and analogy。 While it’s illuminating to see the problems faced in contemporary science， I suppose it's not that surprising， given the drive toward profit and prestige of the professional fields。 And the situation is better in science than in finance。 （And just think of popular social and political discourse！ If knowledge production is fouled up， no wonder the hoi polloi are even more confused。 I wonder if anyone out there is bringing all of this together as one meta-epistemic crisis。） Anyway， this book is divided by sections into expositions on the problems of fraud， bias， negligence and hype， all of which contain cases outstanding and horrid enough to make the reader suspect that Ritchie could be guilty of hyping himself。 I was not aware that Philip Zimbardo's Stanford Prison Experiment has been thoroughly discredited， which just goes to show that retracted science often lives on， in zombie form。 But he does anticipate this response and qualifies his narrative with assurances that he is only concentrating on the most salient examples， and he covers measures currently being undertaken to combat fraud， bias， negligence and hype， so I don’t think he is over-hyping his case here。 The social nature of science led us to the replication crisis of the last decade， affecting all fields of science and decimating some， like psychology。 （A chart of fields affected by inability to replicate results would have been nice here。 Ritchie is potentially guilty of overgeneralizing。）Fraud is what seems to have drawn the most attention in the book so far， and it does occupy a front-and-center position。 Science is a social construct， Ritchie begins， with many players eager to impress each other with salient and positive results of clear originality， so it’s natural that some voices would use false data to gain positive attention， even as the larger community of thinkers has been collegial and trusting， giving fraudsters an environment in which to thrive。 I really don’t think there is any disputing this reasoning， and more stringent skepticism toward results will have to evolve along with reforms to paper publishing， as Ritchie argues in the final portions of the book。 Websites like Retraction watch profile dramatic， even hair-raising cases， but also serve as evidence that the community of knowledge production is taking steps to address the problem。 （One wonders what steps meta-analysts in China might be taking。 And it’s important to note that the argument that political conditions in China encourage fraud is a conjecture without data of any kind to support it， though it sort of makes sense。）As Ritchie says， bias is even more disturbing than fraud， because the former is not based on malevolent intent， is clearly more widespread， and will be much tougher to fix。 Publication bias against null results masks hypotheses proven wrong from view， which means they might be tried out again， and again。 （I wonder if there is solid evidence that that happens。 I notice that some critics still affirm that it’s better not to publish negative results。） “P-hacking” is a form of confirmation bias that has scientists of all stripes massaging their data to strengthen positive results， with tremendous implications for anyone looking to use or build on such work — refer here studies of the effect of priming， as well as of certain medical treatments。 String theory is susceptible to a similar form of confirmation hypothesis， according to the book Lost in Math: How Beauty Leads Physics Astray。 （Again， one would like to see a chart of the hypothetical prevalance of such bias across fields。） Negligence was a particularly shocking chapter。 First there is the fact that the data sets of many labs full of errors as revealed by GRIM testing， which is bad enough， but the more dramatic episode of negligence involves ‘cell lines，’ maintained living tissue samples used in medical research。 These are often simply mislabeled， with the result that work on bone cancer might accidentally use a lung cancer line， or else the very species the tissue was taken from is mistaken。 This is such a common problem， estimates are that it affects tens of thousands of papers worldwide， and maybe as many as 46 percent of cell lines in China。 Science is obstructed by simple book-keeping mistakes！ The level of negligence here is mind-boggling， and really invites deeper inquiry， I think。 And the chapter ends with a consideration of ‘low powered’ research， meaning sample sizes are too small。 The basic logic is that graduate students and lower-level scholars need to get research published， and so must do smaller studies， but the results might include ‘mirage’ effects such as those observed in the case of candidate gene studies， which hoped to link traits to specific genes， an approach thought promising until larger-scale studies showed effects in smaller studies were imaginary。 （Interesting that Ronald Fisher theorized the polygenic nature of traits in 1918; that’s a case where scientists should have known the history of their own fields。）This last type of negligence brushes up against hype in science。 Science must report and draw attention to its results， so inevitably some will take up language that makes the findings dramatic and important。 Results in nutrition and psychology that feed into the self-help industry seem particularly vulnerable， as Ritchie reports on these discredited books: Mindset: The New Psychology of Success]， Before You Know It: The Unconscious Reasons We Do What We Do， and Why We Sleep: Unlocking the Power of Sleep and Dreams， all of which hype data that is of too small scale or even incorrect and retracted。 NASA has gotten in trouble， of course， and one of the biggest data sets from a study of the effects of the Mediterranean diet deserves skepticism even after corrections thanks fact-checking by John Carlisle， a data hound who emerges as a kind of Lone Ranger against fraud， bias， negligence and hype in this book。 Ritchie’s proposed solutions to upgrade science will not surprise readers of behavioral economics cases: it’s about setting up defaults and incentives so that quality is rewarded， where too often today quantity is rewarded。 To improve data quality， more diverse statistical analysis is called for， but just which and how to implement them is a more difficult question。 The distinction between p-values and Bayesian statistics， for example， was revealing but subtle to me — apparently p-values are “almost always calculated independently of any prior evidence。” In that case， how are p-values calculated？ Looking over the book， it’s not surprising that Ritchie skims over the question， referring us to The Art of Statistics: How to Learn from Data and Understanding Psychology as a Science: An Introduction to Scientific and Statistical Inference。All in all， this is a fantastic addition to what I think we may call a new literature off critical thinking。 Precisely because the world is so full of thorny problems， we need to step up our game when it comes to establishing good data and sound analysis based on un-hyped reasoning。 。。。more
Prof。Del
，
Exceptional book。 An honest potrayal of why we are in such a mess right now and what to do to put things right。 Science is not perfect and scientists are certainly not。 Maybe someone in government might want to read it。
Michiel Mennen
，
Although a number of examples in this book of scientific malfeasance were well known to me， seeing all the issues in today's scientific world gathered in one book makes for quite an impactful read。 Recommended！ Although a number of examples in this book of scientific malfeasance were well known to me， seeing all the issues in today's scientific world gathered in one book makes for quite an impactful read。 Recommended！ 。。。more
inexpensive
，
Excellent book about the corruption of science！ A must read for the scientifically minded
Robert Foley
，
Excellent synopsis of the major problems facing scientific research in academia。 Some good concrete proposals for moving forward。 Ritchie does an excellent job of explaining technical issues an accessible， succinct， and engaging way without compromising on accuracy。
Jeffrey Fredericks
，
I found this book to be refreshingly informative。 It made me take the stance I will not simply believe any scientific study out of hand。 Scientific studies are meant to be scrutinized， questions， and overturned， if necessary。 That is the very essence of science。 Job well done Mr。 Ritchie！
Sven Gerst
，
Accessible， rigorous， and witty introduction to meta-science。 Our social trust in Science（tm） needs to be grounded in the scientific method as well as the sociology of how science is practiced by its practitioners。 And especially the latter has gone astray—turning respectable fields such as psychology into a laughing stock。 I am not fully convinced by the “Open Science” approach that the author advances; but in times of eroding trust in actual knowledge， stating the problem as clearly as this bo Accessible， rigorous， and witty introduction to meta-science。 Our social trust in Science（tm） needs to be grounded in the scientific method as well as the sociology of how science is practiced by its practitioners。 And especially the latter has gone astray—turning respectable fields such as psychology into a laughing stock。 I am not fully convinced by the “Open Science” approach that the author advances; but in times of eroding trust in actual knowledge， stating the problem as clearly as this book does （not that it would be somewhat new though）， is a great way to start。 。。。more
Jonathan
，
8/10“Science， the discipline in which we should find the harshest skepticism， the most pin-sharp rationality and the hardest-headed empiricism， has become home to a dizzying array of incompetence， delusion， lies and self-deception。”We have long been facing a replication crisis that seems to have no easy resolution。 The funding for large scale studies， and duplicative studies simply isn't there， and until it is， neither will the research。 It is more difficult to prove fraud then to perpetrate it; 8/10“Science， the discipline in which we should find the harshest skepticism， the most pin-sharp rationality and the hardest-headed empiricism， has become home to a dizzying array of incompetence， delusion， lies and self-deception。”We have long been facing a replication crisis that seems to have no easy resolution。 The funding for large scale studies， and duplicative studies simply isn't there， and until it is， neither will the research。 It is more difficult to prove fraud then to perpetrate it; and when funding or prestigious jobs might depending on publish new results， and only significant results are published， then we have a system that incentives finding 'results'， regardless of certainty。 ‘Published’ and ‘True’ are not synonyms。 If you need some convincing on this front， look at the following: 1 in 25 studies are outright fraudulent， with a higher number unreproducible。 Two percent of scientists are responsible for 25 percent of all retracted papers。 Only 1 in 1000 papers are retracted。 Scientists do not typically publish null results。 Positive result studies make up nearly 90% of studies。 This causes null result studies to be duplicated potentially thousands of times as scientists are unaware it's already been done because no one has bothered to publish it。 The granularity-related inconsistency of means （GRIM） test is a simple statistical test used to identify inconsistencies in the analysis of data sets。 "Brown and Heathers applied the test to 260 articles published in Psychological Science， Journal of Experimental Psychology: General， and Journal of Personality and Social Psychology。 Of these articles， 71 were amenable to GRIM test analysis; 36 of these contained at least one impossible value and 16 contained multiple impossible values。" That for me is almost enough to justify restarting the whole field from scratch。 How can we rely on such tenuous data？Ritchie points out a major problem with scientific studies， showing the rotting underbelly of the gilded edifice that has lifted society into modernity。 We must look at the problems head on， and attempt to remove the bias from researchers。 We need to focus on practical significance， rather than statistical significance。 Just include a confidence interval instead of a p value。 This will be a difficult journey， and it starts with addressing the issue， something Ritchie has done admirably。 。。。more
Rafael Ramirez
，