<\/p>\n
\u00d6\u011frencilerin istatistikte s\u0131kl\u0131kla kar\u0131\u015ft\u0131rd\u0131\u011f\u0131 iki terim p de\u011feri<\/strong> ve alfad\u0131r<\/strong> .<\/span><\/p>\n Her iki terim de, bir hipotezi reddetmek veya reddetmek i\u00e7in kulland\u0131\u011f\u0131m\u0131z resmi istatistiksel testler olan hipotez testinde<\/a> kullan\u0131l\u0131r.<\/span><\/p>\n \u00d6rne\u011fin, yeni bir hap\u0131n hastalar\u0131n kan bas\u0131nc\u0131n\u0131 mevcut standart hapa g\u00f6re daha fazla d\u00fc\u015f\u00fcrd\u00fc\u011f\u00fcn\u00fc varsayd\u0131\u011f\u0131m\u0131z\u0131 varsayal\u0131m.<\/span><\/p>\n Bunu test etmek i\u00e7in a\u015fa\u011f\u0131daki bo\u015f ve alternatif hipotezleri tan\u0131mlad\u0131\u011f\u0131m\u0131z bir hipotez testi yapabiliriz:<\/span><\/p>\n Bo\u015f hipotez:<\/strong> Yeni hap ile standart hap aras\u0131nda hi\u00e7bir fark yoktur.<\/span><\/p>\n Alternatif hipotez:<\/strong> Yeni hap ile standart hap aras\u0131nda fark var<\/em> .<\/span><\/p>\n S\u0131f\u0131r hipotezinin do\u011fru oldu\u011funu varsayarsak, testin p de\u011feri<\/strong> bize en az\u0131ndan \u00f6rnek verilerde ger\u00e7ekte g\u00f6zlemledi\u011fimiz kadar b\u00fcy\u00fck bir etki elde etme olas\u0131l\u0131\u011f\u0131n\u0131 s\u00f6yler.<\/span><\/p>\n \u00d6rne\u011fin, hipotez testi i\u00e7in p de\u011ferinin 0,02 oldu\u011funu buldu\u011fumuzu varsayal\u0131m.<\/span><\/p>\n Bu p de\u011ferini \u015fu \u015fekilde yorumlayabiliriz: E\u011fer yeni hap ile standart hap aras\u0131nda ger\u00e7ekten bir fark yoksa, bu hipotez testini uygulad\u0131\u011f\u0131m\u0131z seferlerin %2’sinde veri \u00f6rne\u011finde g\u00f6zlemlenen etkiyi elde edece\u011fiz veya daha fazlas\u0131, sadece rastgele \u00f6rnekleme hatas\u0131 nedeniyle.<\/span><\/p>\n Bu bize, yeni hap ile standart hap aras\u0131nda ger\u00e7ekten bir fark olmasayd\u0131, ger\u00e7ekte yapt\u0131\u011f\u0131m\u0131z veri \u00f6rneklerini elde etmenin olduk\u00e7a nadir olaca\u011f\u0131n\u0131 s\u00f6yl\u00fcyor.<\/span><\/p>\n Dolay\u0131s\u0131yla s\u0131f\u0131r hipotezini reddetme ve yeni hap ile standart hap aras\u0131nda bir fark oldu\u011fu<\/em> sonucuna varma e\u011filiminde oluruz.<\/span><\/p>\n Ancak p de\u011ferimizin s\u0131f\u0131r hipotezini reddedecek kadar d\u00fc\u015f\u00fck olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in hangi e\u015fi\u011fi kullanmal\u0131y\u0131z?<\/span><\/p>\n \u0130\u015fte alfan\u0131n devreye girdi\u011fi yer buras\u0131!<\/span><\/p>\n Bir hipotez testinin alfa d\u00fczeyi<\/strong> , p de\u011ferimizin s\u0131f\u0131r hipotezini reddedecek kadar d\u00fc\u015f\u00fck olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in kulland\u0131\u011f\u0131m\u0131z e\u015fiktir. Genellikle 0,05’e ayarlan\u0131r, ancak bazen 0,01 veya 0,10’a da ayarlan\u0131r.<\/span><\/p>\n \u00d6rne\u011fin, bir hipotez testinin alfa d\u00fczeyini 0,05’e ayarlay\u0131p 0,02’lik bir p de\u011feri elde edersek, p de\u011feri alfa d\u00fczeyinden k\u00fc\u00e7\u00fck oldu\u011fundan s\u0131f\u0131r hipotezini reddederiz. B\u00f6ylece alternatif hipotezin do\u011fru oldu\u011funu s\u00f6ylemek i\u00e7in yeterli kan\u0131t\u0131m\u0131z oldu\u011fu sonucuna var\u0131r\u0131z.<\/span><\/p>\n Alfa seviyesinin ayn\u0131 zamanda ger\u00e7ek bir s\u0131f\u0131r hipotezinin yanl\u0131\u015fl\u0131kla reddedilme olas\u0131l\u0131\u011f\u0131n\u0131 da tan\u0131mlad\u0131\u011f\u0131n\u0131 belirtmek \u00f6nemlidir.<\/span><\/p>\n \u00d6rne\u011fin, yeni bir hap ile mevcut hap aras\u0131nda kan bas\u0131nc\u0131ndaki ortalama d\u00fc\u015f\u00fc\u015f a\u00e7\u0131s\u0131ndan bir fark olup olmad\u0131\u011f\u0131n\u0131 test etmek istedi\u011fimizi varsayal\u0131m. Ve iki hap aras\u0131nda hi\u00e7bir<\/em> fark olmad\u0131\u011f\u0131n\u0131 varsayal\u0131m.<\/span><\/p>\n Bir hipotez testinin alfa d\u00fczeyini 0,05 olarak ayarlarsak, bu, hipotez test s\u00fcrecini birka\u00e7 kez tekrarlarsak, vakalar\u0131n yakla\u015f\u0131k %5’inde s\u0131f\u0131r hipotezini yanl\u0131\u015fl\u0131kla reddetmeyi bekleyece\u011fimiz anlam\u0131na gelir. testler.<\/span><\/p>\n Daha \u00f6nce de belirtildi\u011fi gibi, bir hipotez testinin alfa d\u00fczeyi i\u00e7in en yayg\u0131n se\u00e7im 0,05’tir. Ancak hatal\u0131 \u00e7\u0131kar\u0131mlar\u0131n ciddi sonu\u00e7lara yol a\u00e7t\u0131\u011f\u0131 baz\u0131 durumlarda alfa d\u00fczeyini daha da a\u015fa\u011f\u0131ya, belki 0,01’e ayarlayabiliriz.<\/span><\/p>\n \u00d6rne\u011fin t\u0131p alan\u0131nda ara\u015ft\u0131rmac\u0131lar\u0131n alfa d\u00fczeyini 0,01 olarak ayarlamas\u0131 yayg\u0131nd\u0131r \u00e7\u00fcnk\u00fc bir hipotez testinin sonu\u00e7lar\u0131n\u0131n g\u00fcvenilir oldu\u011fundan emin olmak isterler.<\/span><\/p>\n Tersine, pazarlama gibi alanlarda alfa d\u00fczeyini 0,10 gibi daha y\u00fckse\u011fe ayarlamak daha yayg\u0131n olabilir \u00e7\u00fcnk\u00fc hata yapman\u0131n sonu\u00e7lar\u0131 ne \u00f6l\u00fcm ne de \u00f6l\u00fcmd\u00fcr.<\/span><\/p>\n Bir testin alfa d\u00fczeyinin art\u0131r\u0131lmas\u0131n\u0131n anlaml\u0131 bir test sonucu bulma \u015fans\u0131n\u0131 art\u0131raca\u011f\u0131, ancak ayn\u0131 zamanda ger\u00e7ek bir s\u0131f\u0131r hipotezini yanl\u0131\u015fl\u0131kla reddetme \u015fans\u0131m\u0131z\u0131 da art\u0131raca\u011f\u0131 unutulmamal\u0131d\u0131r.<\/span><\/p>\n \u0130\u015fte bu makalede \u00f6\u011frendiklerimiz:<\/span><\/p>\n 1.<\/strong> Bir p de\u011feri<\/strong> bize en az\u0131ndan \u00f6rnek verilerde ger\u00e7ekte g\u00f6zlemledi\u011fimiz kadar b\u00fcy\u00fck bir etki elde etme olas\u0131l\u0131\u011f\u0131n\u0131 s\u00f6yler.<\/span><\/p>\n 2.<\/strong> Alfa d\u00fczeyi,<\/strong> do\u011fru bir s\u0131f\u0131r hipotezinin yanl\u0131\u015fl\u0131kla reddedilme olas\u0131l\u0131\u011f\u0131d\u0131r.<\/span><\/p>\n 3.<\/strong> Bir hipotez testinin p de\u011feri alfa seviyesinden k\u00fc\u00e7\u00fckse s\u0131f\u0131r hipotezini reddedebiliriz.<\/span><\/p>\n 4.<\/strong> Bir testin alfa d\u00fczeyini art\u0131rmak, anlaml\u0131 bir test sonucu bulma \u015fans\u0131m\u0131z\u0131 art\u0131r\u0131r, ancak ayn\u0131 zamanda ger\u00e7ek bir s\u0131f\u0131r hipotezini yanl\u0131\u015fl\u0131kla reddetme \u015fans\u0131m\u0131z\u0131 da art\u0131r\u0131r.<\/span><\/p>\n Hipotez Testine Giri\u015f<\/a> \u00d6\u011frencilerin istatistikte s\u0131kl\u0131kla kar\u0131\u015ft\u0131rd\u0131\u011f\u0131 iki terim p de\u011feri ve alfad\u0131r . Her iki terim de, bir hipotezi reddetmek veya reddetmek i\u00e7in kulland\u0131\u011f\u0131m\u0131z resmi istatistiksel testler olan hipotez testinde kullan\u0131l\u0131r. \u00d6rne\u011fin, yeni bir hap\u0131n hastalar\u0131n kan bas\u0131nc\u0131n\u0131 mevcut standart hapa g\u00f6re daha fazla d\u00fc\u015f\u00fcrd\u00fc\u011f\u00fcn\u00fc varsayd\u0131\u011f\u0131m\u0131z\u0131 varsayal\u0131m. Bunu test etmek i\u00e7in a\u015fa\u011f\u0131daki bo\u015f ve alternatif hipotezleri tan\u0131mlad\u0131\u011f\u0131m\u0131z […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[],"class_list":["post-1520","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n Alfa seviyesi<\/strong><\/span><\/h3>\n
Alfa seviyesi nas\u0131l se\u00e7ilir<\/strong><\/span><\/h3>\n
\u00d6zet:<\/strong><\/span><\/h3>\n
Ek kaynaklar<\/strong><\/span><\/h3>\n
S\u0131f\u0131r Hipotezi Nas\u0131l Yaz\u0131l\u0131r (5 \u00d6rnek)<\/a>
Sol ve Sol Nas\u0131l Tan\u0131mlan\u0131r? Do\u011fru Test<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"