Self-Improvement Myths, Checked at the Source

Every claim here is checked against its original research — graded, sourced, and measured against what the top of Google still gets wrong, with our own numbers from reading those ranking pages. Titles link to the full audit; filter by category or search below. The procedure behind every grade is written down separately.

Going deeper on one topic: Procrastination · Attention · Habits · Energy · Books. Everything published on the site, filterable by kind and topic, is in the full index.

CConfirmed

PPartial

HUntested / hypothesis

RRefuted

—Queued, not yet graded

Claim	Grade	Where it comes from	Evidence
Mindset & Self-BeliefDoes Grit Predict Success Better Than Talent?Grit, defined as passion and perseverance for long-term goals, predicts success better than talent or IQ. Grit does predict outcomes, but weakly (rho = .18 to .19 in the two largest syntheses) and largely because it measures conscientiousness under a new name; it beats measured ability only for the narrow outcome of not quitting a gruelling selection course, while in a nationally representative American sample intelligence explained 48 to 90 times more of the variance in educational success. Still on Google41 of 87 top-ranking pages state that grit predicts success better than talent or IQ, 17 hedge, 17 call the claim overstated or false and 12 avoid the comparison, while 20 of 87 mention the Credé 2017 meta-analysis and 13 say grit is largely conscientiousness under another name Checked Sep 5, 2026	P	Angela Duckworth, then a doctoral student at the University of Pennsylvania, with Christopher Peterson (University of Michigan) and West Point officers Michael Matthews and Dennis Kelly, published the Grit Scale in the Journal of Personality and Social Psychology in 2007. The abstract reported that grit accounted for an average of 4% of the variance in success outcomes and showed incremental validity over IQ and conscientiousness; the West Point finding, that grit beat the academy's own Whole Candidate Score at predicting who completed summer training, became the headline. The 2013 TED talk and the 2016 book Grit carried the stronger version, that grit rather than social intelligence, looks, health or IQ is what separates achievers.	Why it spreadDuckworth's 2013 TED talk has been viewed more than 38 million times on TED's own site, the 2016 book reached number two on the New York Times bestseller list, and a memorable but misread sentence from Duckworth and Quinn (2009) told readers that gritty cadets were '99% more likely' to finish summer training, which sounds like a doubled chance rather than a doubled odds ratio in a group where roughly 94% finished anyway. What actually holdsCrede, Tynan and Harms (2017) pooled 584 effect sizes from 88 samples and 66,807 people: grit correlated with academic performance at rho = .18 and with conscientiousness at rho = .84, higher than two different conscientiousness questionnaires correlate with each other, and overall grit added no incremental variance to academic performance after conscientiousness was controlled. Rimfeld and colleagues (2016), using 4,642 UK sixteen-year-olds, found grit added 0.5% to the prediction of exam grades and shared a genetic correlation of 0.86 with conscientiousness. Zisman and Ganzach (2021), using a nationally representative American sample rather than a selected one, found intelligence contributed 48 to 90 times more than grit to educational success.
Motivation & ProductivityAre You 42% More Likely to Achieve a Written Goal?Writing your goals down makes you 42% more likely to achieve them, from a study by Dr Gail Matthews at Dominican University of California. The study is real and randomised but small, four weeks long, self-rated and never published in a journal, and 42% is not a change in likelihood: it is the ratio of two average self-ratings, 6.08 against 4.28. Still on Google49 of 95 top-ranking pages state that writing goals down makes you 42% more likely to achieve them, 2 qualify it and 1 calls it overstated, while 6 of 95 say the study was a conference talk or was never published and 10 name the 149 completers or the self-rated outcome Checked Sep 5, 2026	P	Gail Matthews, then a professor in the Department of Psychology at Dominican University of California, presented the study as a podium talk at the 87th Convention of the Western Psychological Association in Vancouver in 2007, and again in May 2015 at the Ninth Annual International Conference of the Psychology Research Unit of the Athens Institute for Education and Research. Her own three-page Goals Research Summary contains no percentage attached to written goals at all: the only percent signs in it belong to the Yale and Harvard legend she was debunking. The figure 42 first appears in wide circulation on a page by the American author and former publishing executive Michael Hyatt, absent from web archive captures of 1 November 2013 and present in the capture of 6 January 2014, more than a year before Dominican University issued its own press release.	Why it spreadThe number arrived as the honest replacement for the Harvard and Yale goal studies that no archive could produce, so it carries the authority of a real professor, a real university and a real control group. It travelled through Michael Hyatt's goal-setting posts, Dominican's own January 2015 press release and a 2019 CNBC column, and 49 of the 95 readable top-ranking pages we opened still state it as an established finding. What actually holdsThe 42 is arithmetic on the study's own bar chart: mean self-rated achievement of 6.08 in the written-goal group divided by 4.28 in the unwritten-goal group gives 1.4206, a ratio of average ratings rather than a change in odds. The outcome was a self-rating filled in by the participant after four weeks with no independent check, 118 of the 267 starters never returned, and the summary reports no rating scale, no standard deviations, no confidence intervals and no per-group numbers, with roughly thirty people per condition. Group 3, which wrote goals plus action commitments, scored 5.08, a point below written goals alone. Searches on 5 September 2026 of Crossref, OpenAlex, PubMed and Europe PMC return no journal version: the one live identifier, DOI 10.1037/e656862007-001, resolves to the American Psychological Association's PsycEXTRA collection and is typed dataset. The well-measured literature puts the size of the real effect far lower: d = .34 for goal setting across 384 effect sizes and d+ = 0.40 for monitoring progress across 138 experiments.
Mindset & Self-BeliefDoes Power Posing Work? What Fifteen Years of Replications FoundStanding in a "power pose" for two minutes raises testosterone, lowers cortisol and makes you act bolder. Preregistered replications kept the feeling of power and lost everything else: the hormones and the behaviour did not follow, and the first author of the original study has publicly said she no longer believes the effect is real. Still on Google39 of 99 top-ranking pages still present the hormone or performance effect as established, 17 hedge it and 38 say it failed, while 18 of 99 name the first author's public withdrawal and 20 cite the 2015 Ranehill replication Checked Sep 5, 2026	P	Carney, Cuddy & Yap, 2010, Psychological Science: 42 students held two poses for one minute each, gave saliva before and about 17 minutes after, and took one $2 gamble. The paper reported higher testosterone, lower cortisol, more risk-taking and more felt power in the expansive group, and called the implications "real-world, actionable".	Why it spreadAmy Cuddy's 2012 TED talk carried it further than any paper could: 77,356,075 plays on ted.com as of 5 September 2026, plus a bestselling book and a decade of interview advice built on the two-minute ritual. What actually holdsRanehill et al., 2015 ran the design on 200 people with a blinded experimenter: felt power replicated (d = 0.344, p = .017) while testosterone (d = -0.200), cortisol (d = -0.157) and risk-taking (d = -0.176) did not. Seven preregistered studies in Comprehensive Results in Social Psychology found "virtually zero effect" on behavioural or hormonal measures, and Simmons & Simonsohn's p-curve of the 33 supporting studies found no evidential value.
Self-Control & WillpowerDo Hungry Judges Really Deny Parole?Judges grant parole about 65% of the time right after a meal break and almost 0% just before one, because deciding depletes a mental resource. The numbers are quoted correctly from a real dataset, but the causal half is refuted: case order was not random, and a simulated judge who never gets hungry reproduces most of the same curve. Still on Google44 of 97 top-ranking pages repeat the 65-to-zero parole pattern as settled evidence of decision fatigue and 31 of those present the numbers as proof with no hedge at all, while 32 of 97 mention the case-ordering critique by Weinshall-Margel and Shapard or Glöckner's simulation, 16 add a caveat, 20 say the causal story fails and 17 leave the claim unexamined Checked Sep 5, 2026	P	Shai Danziger of Ben Gurion University of the Negev, Jonathan Levav of Columbia Business School and Liora Avnaim-Pesso of Ben Gurion published 1,112 parole rulings by eight Israeli judges over 50 hearing days in PNAS in April 2011, reporting that favourable rulings drop gradually from about 65% to nearly zero within each decision session and return abruptly to about 65% after a food break. The depletion explanation was a speculation in their discussion, not a measurement: the paper states it has no direct measure of the judges' mental resources and that mood was never recorded.	Why it spreadThe graph is vivid, the moral is easy, and the caricature it appears to prove is older than the study itself. The paper was edited by Daniel Kahneman, covered by Nature News, Scientific American, ABC News, NBC News and National Geographic within days, and has since been repeated in willpower books and productivity guides as the field evidence for decision fatigue. What actually holdsKeren Weinshall-Margel of the Israeli Courts Research Division and John Shapard, formerly of the Federal Judicial Center, collected 227 further decisions and interviewed three attorneys, a panel judge and five prison and court staff: the board clears one prison before breaking, unrepresented prisoners are heard last, and they prevailed 15% of the time against 35% for prisoners with counsel. Andreas Glöckner then simulated a judge with no hunger and no fatigue who avoids starting a long case before a break, and because approvals took 7.37 minutes against 5.21 for refusals, that judge produces nearly the same curve, with the authors' own censoring of the last 5% of each session deepening it further. The mechanism invoked, ego depletion, came out at d = 0.04 with a confidence interval of -0.07 to 0.15 across 23 laboratories and 2,141 people, and in nearly 100,000 American arraignment decisions release rates fall about six percentage points in the pre-lunch and pre-dinner sessions only and do not change after a meal.
Energy, Sleep & Body ClockDo Standing Desks Improve Productivity?A standing desk makes you more focused, more productive and healthier. Sitting time does fall by roughly an hour or more a workday, but the focus, productivity and health-outcome parts of the claim are not supported by the trials that tested them. Still on Google61 of 97 top-ranking pages present better focus or productivity as an established benefit of a standing desk, 14 hedge it, 16 argue against it, 22 repeat the Texas A&M call-centre figure of about 45 percent more productive, and 15 cite the Cochrane review, 12 of them only for sitting time or the low certainty of the evidence Checked Sep 5, 2026	P	A Texas A&M Health Science Center press release of 25 May 2016 announced that call-centre staff with stand-capable desks were about 46 percent more productive. The underlying paper, Garrett et al. (2016), compared six months of existing company records for 167 employees across two job categories, with no randomisation, and its own abstract asks for prospective studies. Standing-desk marketing has repeated the figure ever since.	Why it spreadThe number sounded objective because it came from phone-system logs rather than a questionnaire, and it arrived during the wave of alarm about sitting. Sixty-one of ninety-seven top-ranking pages still present the focus or productivity gain as an established benefit. What actually holdsCochrane 2018 pooled three studies of self-rated work performance: +0.35 points on a 1-10 scale, 95% CI -0.1 to 0.8. The three-arm SMART Work and Life trial (756 staff, 2022) isolated the desk from the coaching around it and found no noticeable difference in job performance. A 2023 meta-analysis of 12 experiments and 603 adults found Hedges' g = 0.13, 95% CI -0.04 to 0.29, for the Stroop effect sitting versus standing.
Motivation & ProductivityWas There Ever a Harvard or Yale Study on Written Goals?A study of the Harvard (or Yale) class of 1953 found that the 3% who wrote down their goals were worth more than the other 97% combined twenty years later. No such study exists: Yale searched its own records and alumni archives and found nothing, and the citation trail closes in a circle among three motivational speakers. Still on Google53 of 124 top-ranking pages present the Harvard 1979 or Yale 1953 goal study as a real finding with no hedge, 10 more print its numbers with a footnote of doubt and 46 say it never happened, while 27 of 124 cite the 1996 Fast Company investigation that traced the citation loop Checked Sep 5, 2026	R	Lawrence Tabak traced the Yale version in Fast Company in December 1996 through Jay Rifenbary's 1995 book No Excuse!, page 200 of the 34th printing of Tony Robbins's Unlimited Power, and Zig Ziglar's video Goals, Setting and Achieving Them on Schedule; asked for sources, Robbins's company pointed to Brian Tracy, Tracy pointed to Ziglar and Ziglar's office pointed back to Robbins. Yale University Library traces one outbreak to a September 1991 Success magazine article by Ken Thuerback, then president of Alpine Log Homes, whose source was a Ziglar audio tape and whose ultimate source Ziglar could not name. The Harvard 1979 MBA variant is credited to Mark McCormack's What They Don't Teach You at Harvard Business School, first published in 1984 per the Open Library edition list, five years before the 1989 follow-up it is said to describe.	Why it stuckThe numbers are memorable and flattering, and they were carried by the three biggest names in motivational speaking of the 1980s and 1990s. Brian Tracy's answer when Fast Company put Yale's finding to him was: Heard this story originally from Zig Ziglar, if it's not true it should be. Yale's own archivist put it as: it's just too good not to be true. How it fell apartYale research associate Beverly Waters ran an exhaustive search of the Yale alumni archives and reported that the study did not take place; Silas Spengler, secretary of the class of 1953 since graduation, had never written personal goals and neither he nor his classmates had ever been surveyed; Yale University Library's standing answer states that no goals study of the class of 1953 occurred, after administrators were consulted and the records of various offices examined. Searches of OpenAlex, Europe PMC, PubMed and Crossref on 5 September 2026 return no matching work: zero hits for the phrase pairs written goals plus class of 1953, goals study plus class of 1953, and written goals plus Harvard plus MBA, with only an unrelated 2008 abstract book and a reunion photo caption surfacing as noise.
Mindset & Self-BeliefMaslow Never Drew the PyramidMaslow's pyramid shows five needs that must be satisfied in order. Maslow never drew a pyramid, and the sequence fails: across 123 countries the benefit of meeting any given need was largely independent of whether the others were met. Still on Google20 of 40 pages addressing the pyramid present it as Maslow's own with a sequential order, including MasterClass, Simply Psychology, Cleveland Clinic and tutor2u, 10 criticise the theory's evidence base while still crediting him with the diagram and 10 state that he never drew it, among them Scientific American, Wikipedia and the original Academy of Management paper — pooled from 7 search queries and 51 unique pages Checked Aug 29, 2026	R	Charles McDermid, a consulting psychologist, published the first pyramid in Business Horizons in 1960, in an article titled «How Money motivates Men», arguing it could deliver «maximum motivation at the lowest cost». Keith Davis had drawn a stepped triangle in a 1957 management textbook, topped by a suited executive raising a flag; it did not catch on.	Why it stuckThe image is intuitive, memorable and looks good on a slide, which is why textbooks keep the theory while listing its lack of empirical support in the same chapter. Walter Langer's competing needs theory appeared alongside Maslow's in an early management textbook and vanished. It had no picture. How it fell apartBridgman, Cummings & Ballard searched every book and article Maslow published, his published diaries, and boxes of papers at the Maslow archive at the University of Akron, plus pre-1960 psychology textbooks. No pyramid appears anywhere in his work. His 1943 paper explicitly rejects the readings the triangle invites: needs are partially satisfied at once, a lower need need not be filled before a higher emerges, and the order is not fixed. A 1976 review of thirteen studies found only partial support and none for the gratification-activation proposition. Since he never drew it, he also cannot have taken it from the Blackfoot.
Energy, Sleep & Body ClockCryotherapy’s Whole Evidence Base Is 64 Men, and Cochrane Graded It Very Low«Whole-body cryotherapy speeds recovery, cuts soreness and boosts performance.» The Cochrane review found four trials with 64 mostly male participants, graded every outcome 'very low' quality, and found no study comparing the chamber with cold water immersion. Still on Google82 of 251 cryotherapy pages present the chamber as proven for recovery, soreness, inflammation or fat loss with no caveat, 59 promote it while flagging thin evidence, 60 actively correct it, only 8 note that no trial compared the chamber with a cold bath and just 38 distinguish a chamber from the head-out nitrogen cabin most studios actually sell Checked Aug 30, 2026	P	Whole-body cryotherapy came out of Japanese rheumatology in the late 1970s and spread through European sports medicine, arriving in consumer recovery studios on the back of elite-athlete endorsement rather than any regulatory clearance.	Why it spreadExtreme cold produces a vivid immediate sensation and a plausible anti-inflammatory mechanism, and visible use by professional footballers and sprinters substitutes for evidence in the consumer market. What actually holdsCostello et al. (2015, Cochrane) found only four trials totalling 64 participants, all at high risk of bias, all outcomes graded very low, with confidence intervals crossing zero at three of four time points and no active surveillance of adverse events. Feng et al. (2024) pooled 499 participants and concluded cryostimulation may have no significant influence on exercise performance. Legrand et al. (2023) further show that most consumer studios use head-out nitrogen cabins rather than the breathable-air chambers used in the research. Roberts et al. (2015) found regular post-training cold immersion produced smaller 12-week gains in strength and muscle size than active recovery. Liu and Ma (2026) supplied the head-to-head Cochrane lacked, pooling 13 trials and finding cold water immersion beat cryotherapy for soreness within 24 hours.
Self-Control & WillpowerDoes the Marshmallow Test Predict Anything?A four-year-old who can wait for a second marshmallow will get better grades, better SAT scores and a better life. A small association with school achievement is real and keeps reappearing, but it halves in a diverse sample, loses roughly two thirds of what remains once family background and early ability are accounted for, sits almost entirely in the first twenty seconds of waiting, and is absent from nine adult outcomes at age 26. Still on Google28 of 113 top-ranking pages still present the marshmallow test as a predictor of later success without qualification, 29 hedge it and 50 say the link shrinks or vanishes once family background is controlled, while 69 of 113 mention the 2018 Watts replication and 13 of 113 still give the 210-point SAT figure as fact Checked Sep 5, 2026	P	Shoda, Mischel and Peake, 1990, Developmental Psychology. Of 653 children tested at the Bing School at Stanford between 1968 and 1974, 185 returned any follow-up data and parents of 94 reported SAT scores. The celebrated correlations (verbal r = .42, quantitative r = .57) come from the 35 children tested with the treats in view and no strategy suggested; in the three other conditions (n = 33, 14, 12) the correlations were negative and not significant, and the authors reported a 95% CI of .10 to .66 for the verbal figure and warned the coefficients could exaggerate the true association.	Why it spreadDaniel Goleman's Emotional Intelligence (1995) sold about five million copies and, as the historian Michael Staub documents, leaned almost entirely on Mischel's delay experiments to make self-control the master aptitude, adding the claim that anyone could acquire the skill. The often-quoted figure of 210 SAT points appears nowhere in the 1990 paper, which reports correlations only. What actually holdsWatts, Duncan and Quan, 2018, Psychological Science: in 918 children from the NICHD Study of Early Child Care and Youth Development, age-15 achievement was predicted at 0.24 SD bivariate, 0.08 with family-background controls and a nonsignificant 0.05 with 54-month ability controls, with behaviour composites showing no association even bivariate, and the whole gradient collapsing to a threshold at twenty seconds. Sperber, Vandell, Duncan and Watts, 2024, Child Development, followed 702 of the same children to age 26: of nine outcomes, only educational attainment (r = .17) and BMI (r = -.17) showed raw associations and both fell to nonsignificance under controls.
Attention & FocusDoes Music Help You Focus? From the Mozart Effect to Lo-FiMusic helps you focus and get more done; put the right sound on (Mozart, lo-fi, your favourite playlist) and you work better. Music reliably improves how work feels and can speed up dull, repetitive tasks, but for reading, writing and remembering, songs with words cost about as much as a colleague talking, instrumental music lands on zero, and lo-fi has never been shown to beat silence. Still on Google51 of 119 top-ranking pages present music as a proven focus or productivity aid, 42 hedge it and 21 call it a myth or a distraction, while 57 of 119 warn that lyrics hurt reading and only 15 name the 36-student sample behind the Mozart effect Checked Sep 5, 2026	P	Rauscher, Shaw & Ky, 1993, a one-page letter in Nature: 36 undergraduates scored about 8 to 9 spatial-IQ points higher on one Stanford-Binet subtest after ten minutes of a Mozart sonata, an effect the authors said did not last beyond 10 to 15 minutes and that involved no music during any task. Don Campbell trademarked the phrase and published The Mozart Effect in 1997, and in 1998 Georgia's governor Zell Miller sent a classical CD home with every newborn in the state.	Why it spreadBangerter & Heath, 2004 tracked the fifty largest US newspapers: mentions of college students, the only people ever tested, fell from 80 percent of articles in 1994 to about 30 percent by 2000 while mentions of babies rose from zero to 55 percent, and more than 80 percent of 496 people surveyed in California and Arizona had heard of the effect. What actually holdsChabris, 1999 pooled 16 studies (714 people) and got d = 0.09 for Mozart versus silence; Pietschnig, Voracek & Formann, 2010 pooled nearly 40 studies (3,000+ people) and found the same bump after any music (d = 0.38 versus d = 0.37 for Mozart), with larger effects in Rauscher's own lab. For music during work, Kämpfe et al., 2011 (97 studies) found a null on average with reading negative in all eight studies (r = -0.11), and Vasilev, Kirkby & Angele, 2018 (65 studies) found lyrical music cut reading comprehension like intelligible speech while instrumental music did not.