Изворни код странице P-vrednost

{{DISPLAYTITLE:''p''-vrednost}}

U [[statistical hypothesis testing|statističkom testiranju hipoteze]], '''-{''p''}--vrednost''' ({{jez-eng-lat|probability value}}) ili '''vrednost verovatnoće''', za dati [[statistical model|statistički model]] verovatnoća je da će, kada je [[Null hypothesis|nulta hipoteza]] tačna, [[Summary statistics|statistički parametar]] (kao što je apsolutna vrednost prosečne razlike između dve upoređene grupe) biti veći ili jednak [[Realization (probability)|stvarno uočenom rezultatima]].<ref name="ASA">{{cite journal |last1 = Wasserstein |first1 = Ronald L. |last2 = Lazar |first2 = Nicole A. |date=7. 3. 2016 |title = The ASA's Statement on p-Values: Context, Process, and Purpose |journal= The American Statistician |volume = 70 |issue = 2 |pages = 129–133 |doi= 10.1080/00031305.2016.1154108 |url = https://amstat.tandfonline.com/doi/full/10.1080/00031305.2016.1154108 }}</ref> Upotreba -{''p''}--vrednosti u testiranju statističkih hipoteza je uobičajena u mnogim oblastima istraživanja<ref name="Bhattacharya2002">{{cite journal |last1 = Bhattacharya |first1 = Bhaskar |last2 = Habtzghi |first2 = DeSale |year = 2002 |title = Median of the p value under the alternative hypothesis |url = https://archive.org/details/sim_american-statistician_2002-08_56_3/page/202 |journal = The American Statistician |volume = 56 |issue = 3 |pages = 202–6 |doi = 10.1198/000313002146 }}</ref> kao što su [[fizika]], [[ekonomija]], [[finansije]], [[političke nauke]], [[psihologija]],<ref>{{cite journal |last1 = Wetzels | first1 = R. | last2 = Matzke | first2 = D. | last3 = Lee | first3 = M. D. | last4 = Rouder | first4 = J. N. |last5 = Iverson |first5 = G. J. |last6 = Wagenmakers |first6 = E. -J. |doi = 10.1177/1745691611406923 | title = Statistical Evidence in Experimental Psychology: An Empirical Comparison Using 855 t Tests |journal = Perspectives on Psychological Science |volume = 6 |issue = 3 |pages = 291–298 |year = 2011 |pmid =  26168519 |pmc = }}</ref> [[biologija]], [[krivično pravo]], [[kriminologija]] i [[sociologija]].<ref>Babbie, E. (2007). The practice of social research 11th ed. Thomson Wadsworth: Belmont, California.</ref> [[misuse of p-values|Zloupotreba -{''p''}--vrednosti]] je kontroverzna tema u [[metascience|metanauci]].<ref>{{cite web |last1=Ioannidis |first1=John P. A. |last2=Ware |first2=Jennifer J. |last3=Wagenmakers |first3=Eric-Jan |last4=Simonsohn |first4=Uri |last5=Chambers |first5=Christopher D. |last6=Button |first6=Katherine S. |last7=Bishop |first7=Dorothy V. M. |last8=Nosek |first8=Brian A. |last9=Munafò |first9=Marcus R. |title=A manifesto for reproducible science |url=https://www.nature.com/articles/s41562-016-0021 |website=Nature Human Behaviour |accessdate=9. 5. 2019 |pages=0021 |language=en |doi=10.1038/s41562-016-0021 |date=januar 2017}}</ref>

Zakošavanje slova, upotreba velikih slova i deljenja reči variraju. Na primer, [[AMA Manual of Style|AMA stil]] koristi „''P'' vrednost”, [[APA stil]] koristi „''p'' vrednost”, a [[American Statistical Association|Američko statističko udruženje]] koristi „''p''-vrednost”.<ref>[http://magazine.amstat.org/wp-content/uploads/STATTKadmin/style%5B1%5D.pdf ASA House Style]</ref>

== Definicija i interpretacija ==
=== Definicija ===

==== Verovatnoća pod nultom hipotezom dobijanja statistike testa sa realnom vrednošću koja je najmanje ekstremna kao i dobijena ====

Uzmimo u obzir posmatranu statistiku testa <math>t</math> iz nepoznate distribucije <math>T</math>. Tada je ''p''-vrednost <math>p</math> ono što bi bila prethodna verovatnoća da se posmatra statistička vrednost testa barem tako „ekstremna” kao <math>t</math> ako je nulta hipoteza <math>H_0</math> bila istinita. To je:
* <math>p = \Pr(T \geq t \mid H_0)</math> za jednostranu distribuciju test statistike na desni rep,
* <math>p = \Pr(T \leq t \mid H_0)</math> za jednostranu distribuciju test statistike sa leve strane,
* <math>p = 2\min\{\Pr(T \geq t \mid H_0),\Pr(T \leq t \mid H_0)\} </math>  za dvostranu distribuciju test-statistike. Ako je raspodela <math>T</math> simetrična oko nule, onda je <math>p =\Pr(|T| \geq |t| \mid H_0)</math>

=== Tumačenja ===

==== p-vrednost kao statistika za izvođenje testova značajnosti ====

U testu značaja, nulta hipoteza <math>H_0</math> se odbacuje ako je ''p''-vrednost manja ili jednaka unapred definisanoj graničnoj vrednosti <math>\alpha</math>, koja se naziva alfa nivo ili [[statistical significance|nivo značajnosti]]. <math>\alpha</math> nije izvedeno iz podataka, već ga postavlja istraživač pre ispitivanja podataka. <math>\alpha</math> se obično postavlja na 0,05, mada se ponekad koriste niži nivoi za alfa. U 2018. godini, grupa statističara na čelu sa Danijelom Bendžaminom predložila je usvajanje vrednosti od 0,005 kao standardne vrednosti za statističku značajnost širom sveta.<ref name="BenjaminBergerJohannesson2017">{{cite journal | last1 = Benjamin | first1 = Daniel J. | last2 = Berger | first2 = James O. | last3 = Johannesson | first3 = Magnus | last4 = Nosek | first4 = Brian A. | last5 = Wagenmakers | first5 = E.-J. | last6 = Berk | first6 = Richard | last7 = Bollen | first7 = Kenneth A. | last8 = Brembs | first8 = Björn | last9 = Brown | first9 = Lawrence | last10 = Camerer | first10 = Colin | last11 = Cesarini | first11 = David | last12 = Chambers | first12 = Christopher D. | last13 = Clyde | first13 = Merlise | last14 = Cook | first14 = Thomas D. | last15 = De Boeck | first15 = Paul | last16 = Dienes | first16 = Zoltan | last17 = Dreber | first17 = Anna | last18 = Easwaran | first18 = Kenny | last19 = Efferson | first19 = Charles | last20 = Fehr | first20 = Ernst | last21 = Fidler | first21 = Fiona | last22 = Field | first22 = Andy P. | last23 = Forster | first23 = Malcolm | last24 = George | first24 = Edward I. | last25 = Gonzalez | first25 = Richard | last26 = Goodman | first26 = Steven | last27 = Green | first27 = Edwin | last28 = Green | first28 = Donald P. | last29 = Greenwald | first29 = Anthony G. | last30 = Hadfield | first30 = Jarrod D. | last31 = Hedges | first31 = Larry V. | last32 = Held | first32 = Leonhard | last33 = Hua Ho | first33 = Teck | last34 = Hoijtink | first34 = Herbert | last35 = Hruschka | first35 = Daniel J. | last36 = Imai | first36 = Kosuke | last37 = Imbens | first37 = Guido | last38 = Ioannidis | first38 = John P. A. | last39 = Jeon | first39 = Minjeong | last40 = Jones | first40 = James Holland | last41 = Kirchler | first41 = Michael | last42 = Laibson | first42 = David | last43 = List | first43 = John | last44 = Little | first44 = Roderick | last45 = Lupia | first45 = Arthur | last46 = Machery | first46 = Edouard | last47 = Maxwell | first47 = Scott E. | last48 = McCarthy | first48 = Michael | last49 = Moore | first49 = Don A. | last50 = Morgan | first50 = Stephen L. | last51 = Munafó | first51 = Marcus | last52 = Nakagawa | first52 = Shinichi | last53 = Nyhan | first53 = Brendan | last54 = Parker | first54 = Timothy H. | last55 = Pericchi | first55 = Luis | last56 = Perugini | first56 = Marco | last57 = Rouder | first57 = Jeff | last58 = Rousseau | first58 = Judith | last59 = Savalei | first59 = Victoria | last60 = Schönbrodt | first60 = Felix D. | last61 = Sellke | first61 = Thomas | last62 = Sinclair | first62 = Betsy | last63 = Tingley | first63 = Dustin | last64 = Van Zandt | first64 = Trisha | last65 = Vazire | first65 = Simine | last66 = Watts | first66 = Duncan J. | last67 = Winship | first67 = Christopher | last68 = Wolpert | first68 = Robert L. | last69 = Xie | first69 = Yu | last70 = Young | first70 = Cristobal | last71 = Zinman | first71 = Jonathan | last72 = Johnson | first72 = Valen E. | title = Redefine statistical significance | journal = Nature Human Behaviour | date = 1 September 2017 | volume = 2 | issue = 1 | pages = 6–10 | eissn = 2397-3374 | doi = 10.1038/s41562-017-0189-z | pmid = 30980045 ||display-authors= 3 }}</ref>

=== Distribucija ===

Kada je nulta hipoteza tačna, ako ima oblik <math> H_0: \theta = \theta_0</math>, a osnovna slučajna promenljiva je kontinuirana, onda je [[probability distribution|raspodela verovatnoće]] ''p''-vrednosti [[Uniform distribution (continuous)|uniformna]] na intervalu [0,1]. Nasuprot tome, ako je alternativna hipoteza tačna, distribucija zavisi od veličine uzorka i prave vrednosti parametra koji se proučava.<ref name="Bhattacharya2002"/><ref name="Hung1997">{{cite journal | vauthors = Hung HM, O'Neill RT, Bauer P, Köhne K | title = The behavior of the P-value when the alternative hypothesis is true | journal = Biometrics | volume = 53 | issue = 1 | pages = 11–22 | date = March 1997 | pmid = 9147587 | doi = 10.2307/2533093 | type = Submitted manuscript | jstor = 2533093 | url = https://zenodo.org/record/1235121 }}</ref>

Distribucija ''p''-vrednosti za grupu studija se ponekad naziva ''p''-kriva.<ref name="Head2015">{{cite journal | vauthors = Head ML, Holman L, Lanfear R, Kahn AT, Jennions MD | title = The extent and consequences of p-hacking in science | journal = PLOS Biology | volume = 13 | issue = 3 | pages = e1002106 | date = March 2015 | pmid = 25768323 | pmc = 4359000 | doi = 10.1371/journal.pbio.1002106 }}</ref> ''p''-kriva se može koristiti za procenu pouzdanosti naučne literature, kao što je otkrivanje pristrasnosti publikacije ili [[p-hacking|''p''-hakovanje]].<ref name="Head2015"/><ref name="Simonsohn2014">{{cite journal | vauthors = Simonsohn U, Nelson LD, Simmons JP | title = p-Curve and Effect Size: Correcting for Publication Bias Using Only Significant Results | journal = Perspectives on Psychological Science | volume = 9 | issue = 6 | pages = 666–681 | date = November 2014 | pmid = 26186117 | doi = 10.1177/1745691614553988 | s2cid = 39975518 }}</ref>

== Reference ==
{{Reflist}}

== Literatura ==
{{Литература|30em}}
* {{cite journal | last = Pearson | first = Karl | author-link = Karl Pearson | title = On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling | doi = 10.1080/14786440009463897 | journal = Philosophical Magazine |series=Series 5 | volume = 50 | issue = 302 | pages = 157–175 | year = 1900 |url = http://www.economics.soton.ac.uk/staff/aldrich/1900.pdf | pmid =  | pmc = }}
* {{cite journal | last = Elderton | first = William Palin | author-link = William Palin Elderton |title = Tables for Testing the Goodness of Fit of Theory to Observation | url = https://archive.org/details/sim_biometrika_1902-01_1_2/page/155 |doi = 10.1093/biomet/1.2.155 |journal = Biometrika |volume = 1 |issue = 2 | pages = 155–163 | year = 1902 | pmid =  | pmc = }}
* {{cite book |title = Statistical Methods for Research Workers |last=Fisher |first=Ronald |author-link = Ronald Fisher |year=1925 |publisher=Oliver &amp; Boyd |location = Edinburgh, Scotland |isbn = 978-0-05-002170-5 |ref = harv |title-link=Statistical Methods for Research Workers }}
* {{cite book |title = The Design of Experiments |edition=9th |last=Fisher |first=Ronald A. |author-link = Ronald Fisher |origyear=1935 |year=1971 |publisher=Macmillan |isbn = 978-0-02-844690-5 |ref = harv }}
* {{cite book
| last1 = Fisher |first1= R. A. |last2= Yates |first2 = F.
| title = Statistical tables for biological, agricultural and medical research
| url = https://archive.org/details/b32736216 | year = 1938
| location = London, England
}}
* {{cite book |title = The history of statistics : the measurement of uncertainty before 1900 |url = https://archive.org/details/historyofstatist00stig |last=Stigler |first=Stephen M. |author-link = Stephen M. Stigler |year=1986 |publisher=Belknap Press of Harvard University Press |location = Cambridge, Mass |isbn = 978-0-674-40340-6 |ref = harv }}
* {{citation |first1=Raymond |last1=Hubbard |first2=M. J. |last2=Bayarri |author2-link = M. J. Bayarri |url = http://ftp.isds.duke.edu/WorkingPapers/03-26.pdf |title = P Values are not Error Probabilities |date=novembar 2003 |postscript = , a working paper that explains the difference between Fisher's evidential ''p''-value and the Neyman–Pearson Type I error rate ''α''. |url-status=dead |archiveurl = https://web.archive.org/web/20130904000350/http://ftp.isds.duke.edu/WorkingPapers/03-26.pdf |archivedate=4. 9. 2013 }}
* {{cite journal |last1=Hubbard |first1=Raymond |last2=Armstrong |first2=J. Scott |author2-link = J. Scott Armstrong |title = Why We Don't Really Know What Statistical Significance Means: Implications for Educators |doi = 10.1177/0273475306288399 |url=http://hops.wharton.upenn.edu/ideas/pdf/Armstrong/StatisticalSignificance.pdf |journal = Journal of Marketing Education |volume=28 |issue=2 |pages=114–120 |year=2006 |pmid= |pmc= |deadurl=unfit |archiveurl = https://web.archive.org/web/20060518054857/http://hops.wharton.upenn.edu/ideas/pdf/Armstrong/StatisticalSignificance.pdf |archivedate=18. 5. 2006 }}
* {{cite journal |last1 = Hubbard |first1 = Raymond |last2 = Lindsay |first2 = R. Murray |doi = 10.1177/0959354307086923 |title = Why ''P'' Values Are Not a Useful Measure of Evidence in Statistical Significance Testing |journal = Theory & Psychology |volume = 18 |issue = 1 |pages = 69–88 |year = 2008 |url = http://wiki.bio.dtu.dk/~agpe/papers/pval_notuseful.pdf |access-date = 21. 07. 2019 |archive-url = https://web.archive.org/web/20161021014340/http://wiki.bio.dtu.dk/~agpe/papers/pval_notuseful.pdf |archive-date = 21. 10. 2016 |url-status = dead }}
* {{cite journal |last1 = Stigler | first1 = S. |author-link = Stephen Stigler |title = Fisher and the 5% level |doi = 10.1007/s00144-008-0033-3 |journal = Chance | volume = 21 | issue = 4 | page = 12 |date=decembar 2008 |pmid =  |pmc = |ref = harv }}
* {{cite book |title = The Little Handbook of Statistical Practice |first=Gerard E. |last=Dallal |year=2012 |url = http://www.tufts.edu/~gdallal/LHSP.HTM |ref = harv }}
* {{cite journal |last1 = Biau  |first1 = D.J. |last2 = Jolles | first2 = B.M. | last3 = Porcher | first3 = R. |title = P value and the theory of hypothesis testing: an explanation for new researchers |doi = 10.1007/s11999-009-1164-4 |journal = Clin Orthop Relat Res |volume = 463 |issue = 3 |pages = 885–892 |date=mart 2010 |pmid =  19921345 | pmc = 2816758 |ref = harv }}
* {{cite book |last=Reinhart |first=Alex |title=Statistics Done Wrong: The Woefully Complete Guide |publisher=[[No Starch Press]] |url = http://statisticsdonewrong.com |isbn = 978-1593276201 |page = 176 |year = 2015 }}
* {{cite journal | vauthors = Denworth L | title = A Significant Problem: Standard scientific methods are under fire. Will anything change? | journal = [[Scientific American]] | volume = 321 | issue = 4 | date = October 2019 | pages = 62–67 (63) | quote = The use of [[p value|''p'' values]] for nearly a century [since 1925] to determine [[statistical significance]] of [[experiment|experimental]] results has contributed to an illusion of [[certainty]] and [to] [[reproducibility|reproducibility crises]] in many [[science|scientific fields]]. There is growing determination to reform statistical analysis... Some [researchers] suggest changing statistical methods, whereas others would do away with a threshold for defining "significant" results. }} 
* {{cite journal |author-last1=Benjamini |author-first1=Yoav |author-link1=Yoav Benjamini |author-last2=De Veaux |author-first2=Richard D. 
|author-last3=Efron|author-first3=Bradley|author-link3=Bradley Efron|author-last4=Evans|author-first4=Scott
|author-last5=Glickman|author-first5=Mark
|author-last6=Graubard|author-first6=Barry I.
|author-last7=He|author-first7=Xuming
|author-last8=Meng|author-first8=Xiao-Li|author-link8=Xiao-Li Meng
|author-last9=Reid|author-first9=Nancy|author-link9=Nancy Reid
|author-last10=Stigler|author-first10=Stephen M.|author-link10 = Stephen Stigler
|author-last11=Vardeman|author-first11=Stephen B.
|author-last12=Wikle|author-first12=Christopher K.
|author-last13=Wright|author-first13=Tommy
|author-last14=Young|author-first14=Linda J.
|author-last15=Kafadar|author-first15=Karen|author-link15=Karen Kafadar
|journal=Annals of Applied Statistics
|year=2021
|volume=15
|pages=1084–1085
 |title=The ASA President's Task Force Statement on Statistical Significance and Replicability
|issue=3 |doi=10.1214/21-AOAS1501|doi-access=free}}
*{{cite journal | last1 = Benjamin | first1 = Daniel J. | last2 = Berger | first2 = James O. | last3 = Johannesson | first3 = Magnus | last4 = Nosek | first4 = Brian A. | last5 = Wagenmakers | first5 = E.-J. | last6 = Berk | first6 = Richard | last7 = Bollen | first7 = Kenneth A. | last8 = Brembs | first8 = Björn | last9 = Brown | first9 = Lawrence | last10 = Camerer | first10 = Colin | last11 = Cesarini | first11 = David | last12 = Chambers | first12 = Christopher D. | last13 = Clyde | first13 = Merlise | last14 = Cook | first14 = Thomas D. | last15 = De Boeck | first15 = Paul | last16 = Dienes | first16 = Zoltan | last17 = Dreber | first17 = Anna | last18 = Easwaran | first18 = Kenny | last19 = Efferson | first19 = Charles | last20 = Fehr | first20 = Ernst | last21 = Fidler | first21 = Fiona | last22 = Field | first22 = Andy P. | last23 = Forster | first23 = Malcolm | last24 = George | first24 = Edward I. | last25 = Gonzalez | first25 = Richard | last26 = Goodman | first26 = Steven | last27 = Green | first27 = Edwin | last28 = Green | first28 = Donald P. | last29 = Greenwald | first29 = Anthony G. | last30 = Hadfield | first30 = Jarrod D. | last31 = Hedges | first31 = Larry V. | last32 = Held | first32 = Leonhard | last33 = Hua Ho | first33 = Teck | last34 = Hoijtink | first34 = Herbert | last35 = Hruschka | first35 = Daniel J. | last36 = Imai | first36 = Kosuke | last37 = Imbens | first37 = Guido | last38 = Ioannidis | first38 = John P. A. | last39 = Jeon | first39 = Minjeong | last40 = Jones | first40 = James Holland | last41 = Kirchler | first41 = Michael | last42 = Laibson | first42 = David | last43 = List | first43 = John | last44 = Little | first44 = Roderick | last45 = Lupia | first45 = Arthur | last46 = Machery | first46 = Edouard | last47 = Maxwell | first47 = Scott E. | last48 = McCarthy | first48 = Michael | last49 = Moore | first49 = Don A. | last50 = Morgan | first50 = Stephen L. | last51 = Munafó | first51 = Marcus | last52 = Nakagawa | first52 = Shinichi | last53 = Nyhan | first53 = Brendan | last54 = Parker | first54 = Timothy H. | last55 = Pericchi | first55 = Luis | last56 = Perugini | first56 = Marco | last57 = Rouder | first57 = Jeff | last58 = Rousseau | first58 = Judith | last59 = Savalei | first59 = Victoria | last60 = Schönbrodt | first60 = Felix D. | last61 = Sellke | first61 = Thomas | last62 = Sinclair | first62 = Betsy | last63 = Tingley | first63 = Dustin | last64 = Van Zandt | first64 = Trisha | last65 = Vazire | first65 = Simine | last66 = Watts | first66 = Duncan J. | last67 = Winship | first67 = Christopher | last68 = Wolpert | first68 = Robert L. | last69 = Xie | first69 = Yu | last70 = Young | first70 = Cristobal | last71 = Zinman | first71 = Jonathan | last72 = Johnson | first72 = Valen E. | title = Redefine statistical significance | journal = Nature Human Behaviour | date = 1 September 2017 | volume = 2 | issue = 1 | pages = 6–10 | eissn = 2397-3374 | doi = 10.1038/s41562-017-0189-z | pmid = 30980045 |display-authors= 3 |}}
{{Литература крај}}

== Spoljašnje veze ==
{{Commons category-lat|P-value}}
* -{[http://www.danielsoper.com/statcalc/default.aspx#c14 Free online ''p''-values calculators] for various specific tests (chi-square, Fisher's F-test, etc.).}-
* -{[http://www.stat.duke.edu/%7Eberger/p-values.html Understanding ''p''-values], including a Java applet that illustrates how the numerical values of ''p''-values can give quite misleading impressions about the truth or falsity of the hypothesis under test.}-
* {{YouTube-lat|5Z9OIYA8He8|StatQuest: P Values, clearly explained}}
* {{YouTube-lat|UFhJefdVCjE|StatQuest: P-value pitfalls and power calculations}}

{{Authority control-lat}}

[[Категорија:Тестирање статистичких хипотеза]]