A\u015fa\u011f\u0131daki format\u0131 alan 2’ye 2’lik bir tabloda analiz ger\u00e7ekle\u015ftirirken s\u0131kl\u0131kla bir olas\u0131l\u0131k oran\u0131 hesaplar\u0131z:<\/span> <\/p>\n R’de bir olas\u0131l\u0131k oran\u0131 hesaplamak i\u00e7in epitools<\/strong> paketindeki oddsratio()<\/strong> i\u015flevini kullanabiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki \u00f6rnek, bu s\u00f6zdiziminin pratikte nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n Diyelim ki 50 basketbolcu yeni bir antrenman program\u0131 kullan\u0131yor ve 50 oyuncu da eski bir antrenman program\u0131n\u0131 kullan\u0131yor. Program\u0131n sonunda her oyuncuyu belirli bir beceri testini ge\u00e7ip ge\u00e7mediklerini test ediyoruz.<\/span><\/p>\n A\u015fa\u011f\u0131daki tablo, kulland\u0131klar\u0131 programa g\u00f6re ba\u015far\u0131l\u0131 olan ve ba\u015far\u0131s\u0131z olan oyuncular\u0131n say\u0131s\u0131n\u0131 g\u00f6stermektedir:<\/span> <\/p>\n Bir oyuncunun yeni program\u0131 kullanarak beceri testini ge\u00e7me \u015fans\u0131n\u0131 eski program\u0131 kullanarak kar\u015f\u0131la\u015ft\u0131rma \u015fans\u0131n\u0131 kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in bir olas\u0131l\u0131k oran\u0131 hesaplamak istedi\u011fimizi varsayal\u0131m.<\/span><\/p>\n Bu matrisi R’de nas\u0131l olu\u015fturaca\u011f\u0131n\u0131z a\u015fa\u011f\u0131da a\u00e7\u0131klanm\u0131\u015ft\u0131r:<\/span><\/p>\n Epitools<\/strong> paketindeki oddsratio()<\/strong> i\u015flevini kullanarak olas\u0131l\u0131k oran\u0131n\u0131n nas\u0131l hesaplanaca\u011f\u0131 a\u015fa\u011f\u0131da a\u00e7\u0131klanm\u0131\u015ft\u0131r:<\/span><\/p>\n Oran oran\u0131 0,6045506<\/strong> olarak \u00e7\u0131k\u0131yor.<\/span><\/p>\n Bunu, bir oyuncunun yeni program\u0131 kullanarak testi ge\u00e7me \u015fans\u0131n\u0131n, bir oyuncunun testi ge\u00e7me \u015fans\u0131n\u0131n eski program\u0131 kullanarak ge\u00e7me \u015fans\u0131n\u0131n yaln\u0131zca 0,6045506 kat\u0131<\/strong> oldu\u011fu \u015feklinde yorumluyoruz.<\/span><\/p>\n Ba\u015fka bir deyi\u015fle, yeni program\u0131n kullan\u0131lmas\u0131yla bir oyuncunun testi ge\u00e7me \u015fans\u0131 asl\u0131nda yakla\u015f\u0131k %39,6 oran\u0131nda azal\u0131yor.<\/span><\/p>\n Olas\u0131l\u0131k oran\u0131 i\u00e7in a\u015fa\u011f\u0131daki %95 g\u00fcven aral\u0131\u011f\u0131n\u0131 olu\u015fturmak amac\u0131yla sonucun alt<\/strong> ve \u00fcst<\/strong> s\u00fctunlar\u0131ndaki de\u011ferleri de kullanabiliriz:<\/span><\/p>\n Olas\u0131l\u0131k oran\u0131 i\u00e7in %95 g\u00fcven aral\u0131\u011f\u0131: [0,24, 1,48]<\/strong> .<\/span><\/p>\n Yeni ve eski antrenman program\u0131 aras\u0131ndaki ger\u00e7ek odds oran\u0131n\u0131n bu aral\u0131kta yer ald\u0131\u011f\u0131ndan %95 eminiz.<\/span><\/p>\n \u00c7\u0131kt\u0131daki midp.exact<\/strong> s\u00fctunu ayn\u0131 zamanda olas\u0131l\u0131k oran\u0131yla ili\u015fkili p de\u011ferini de g\u00f6r\u00fcnt\u00fcler.<\/span><\/p>\n Bu p de\u011feri 0,271899<\/strong> olarak ortaya \u00e7\u0131k\u0131yor. Bu de\u011fer 0,05’ten k\u00fc\u00e7\u00fck olmad\u0131\u011f\u0131 i\u00e7in odds oran\u0131n\u0131n istatistiksel olarak anlaml\u0131 olmad\u0131\u011f\u0131 sonucuna varabiliriz.<\/span><\/p>\n Yani bir oyuncunun yeni program\u0131 kullanarak ba\u015far\u0131l\u0131 olma \u015fans\u0131n\u0131n eski program\u0131 kullanarak ba\u015far\u0131l\u0131 olma \u015fans\u0131ndan daha d\u00fc\u015f\u00fck oldu\u011funu oran oran\u0131ndan biliyoruz ancak bu \u015fanslar aras\u0131ndaki fark asl\u0131nda istatistiksel olarak anlaml\u0131 de\u011fil.<\/span><\/p>\n A\u015fa\u011f\u0131daki e\u011fitimler oran oranlar\u0131 hakk\u0131nda ek bilgi sa\u011flar:<\/span><\/p>\n Oran oran\u0131 ve g\u00f6receli risk: fark nedir?<\/a> \u0130statistiklerde olas\u0131l\u0131k oran\u0131 bize, bir tedavi grubunda meydana gelen bir olay\u0131n olas\u0131l\u0131\u011f\u0131n\u0131n, bir kontrol grubunda meydana gelen bir olay\u0131n olas\u0131l\u0131\u011f\u0131na oran\u0131n\u0131 belirtir. A\u015fa\u011f\u0131daki format\u0131 alan 2’ye 2’lik bir tabloda analiz ger\u00e7ekle\u015ftirirken s\u0131kl\u0131kla bir olas\u0131l\u0131k oran\u0131 hesaplar\u0131z: R’de bir olas\u0131l\u0131k oran\u0131 hesaplamak i\u00e7in epitools paketindeki oddsratio() i\u015flevini kullanabiliriz. A\u015fa\u011f\u0131daki \u00f6rnek, bu s\u00f6zdiziminin pratikte nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir. […]<\/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-3587","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/cotesratioexcel0.png\")
<\/p>\n \u00d6rnek: R’de bir olas\u0131l\u0131k oran\u0131 hesaplay\u0131n<\/strong><\/span><\/h2>\n
![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/cotes1.png\")
<\/p>\n #create matrix<\/span>\nprogram <- c(' New Program<\/span> ', ' Old Program<\/span> ')\noutcome <- c(' Pass<\/span> ', ' Fail<\/span> ')\ndata <- matrix(c(34, 16, 39, 11), nrow= 2<\/span> , ncol= 2<\/span> , byrow= TRUE<\/span> )\ndimnames(data) <- list(' Program<\/span> '=program, ' Outcome<\/span> '=outcome)\n\n#view matrix\n<\/span>data\n\n Outcome\nProgram Pass Fail\n New Program 34 16\n Old Program 39 11<\/strong><\/pre>\n install. packages<\/span> (' epitools<\/span> ')<\/span>\n\n<\/span>library<\/span> (epitools)\n\n#calculate odds ratio<\/span>\noddsratio(data)\n\n$measure\n odds ratio with 95% CI\nProgram estimate lower upper\n New Program 1.0000000 NA NA\n Old Program 0.6045506 0.2395879 1.480143\n\n$p.value\n two-sided\nProgram midp.exact fisher.exact chi.square\n New Program NA NA NA\n Old Program 0.271899 0.3678219 0.2600686\n\n$correction\n[1] FALSE\n\nattr(,\"method\")\n[1] \u201cmedian-unbiased estimate & mid-p exact CI\u201d\n<\/strong><\/pre>\n Ek kaynaklar<\/strong><\/span><\/h2>\n
Tam K\u0131lavuz: Oran Oranlar\u0131 Nas\u0131l Raporlan\u0131r?<\/a>
Bir olas\u0131l\u0131k oran\u0131 i\u00e7in g\u00fcven aral\u0131\u011f\u0131 nas\u0131l hesaplan\u0131r?<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"