Ancak tekrarlanan \u00f6l\u00e7\u00fcm ANOVA’s\u0131n\u0131 ger\u00e7ekle\u015ftirmeden \u00f6nce a\u015fa\u011f\u0131daki varsay\u0131mlar\u0131n kar\u015f\u0131land\u0131\u011f\u0131ndan emin olmal\u0131y\u0131z:<\/span><\/p>\n 1. Ba\u011f\u0131ms\u0131zl\u0131k:<\/strong> Her g\u00f6zlem ba\u011f\u0131ms\u0131z olmal\u0131d\u0131r.<\/span><\/p>\n 2. Normallik:<\/strong> Yan\u0131t de\u011fi\u015fkeninin da\u011f\u0131l\u0131m\u0131 normal da\u011f\u0131lm\u0131\u015ft\u0131r.<\/span><\/p>\n 3. K\u00fcresellik:<\/strong> \u0130lgili gruplar\u0131n t\u00fcm kombinasyonlar\u0131 aras\u0131ndaki farklar\u0131n varyanslar\u0131 e\u015fit olmal\u0131d\u0131r.<\/span><\/p>\n Bu varsay\u0131mlardan bir veya daha fazlas\u0131 ihlal edilirse, tekrarlanan ANOVA \u00f6l\u00e7\u00fcmlerinin sonu\u00e7lar\u0131 g\u00fcvenilir olmayabilir.<\/span><\/p>\n Bu yaz\u0131da her bir varsay\u0131m i\u00e7in bir a\u00e7\u0131klama, varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131n\u0131n nas\u0131l belirlenece\u011fi ve kar\u015f\u0131lanmamas\u0131 durumunda ne yap\u0131laca\u011f\u0131na dair a\u00e7\u0131klamalar sunuyoruz.<\/span><\/p>\n Tekrarlanan bir ANOVA \u00f6l\u00e7\u00fcm\u00fc, veri k\u00fcmenizdeki her g\u00f6zlemin<\/a> di\u011fer t\u00fcm g\u00f6zlemlerden ba\u011f\u0131ms\u0131z oldu\u011funu varsayar.<\/span><\/p>\n Bu varsay\u0131m\u0131 do\u011frulaman\u0131n en basit yolu, veri k\u00fcmesindeki her bireyin , rastgele \u00f6rnekleme y\u00f6ntemi<\/a> kullan\u0131larak pop\u00fclasyondan<\/a> rastgele \u00f6rneklendi\u011fini do\u011frulamakt\u0131r.<\/span><\/p>\n Rastgele \u00f6rnekleme y\u00f6ntemi kullan\u0131lm\u0131\u015fsa, her g\u00f6zlemin ba\u011f\u0131ms\u0131z oldu\u011fu varsay\u0131labilir.<\/span><\/p>\n Bu varsay\u0131m\u0131n kar\u015f\u0131lanmamas\u0131 ciddi bir sorundur \u00e7\u00fcnk\u00fc her bireyin de\u011ferleri \u015fu veya bu \u015fekilde birbiriyle ili\u015fkili olabilir.<\/span><\/p>\n Genellikle bu senaryodaki tek \u00e7\u00f6z\u00fcm, rastgele \u00f6rnekleme y\u00f6ntemi kullan\u0131larak yeni bir \u00e7al\u0131\u015fma i\u00e7in bireyleri i\u015fe almakt\u0131r.<\/span><\/p>\n Tekrarlanan bir ANOVA \u00f6l\u00e7\u00fcm\u00fc, yan\u0131t de\u011fi\u015fkeninin<\/a> da\u011f\u0131l\u0131m\u0131n\u0131n normal \u015fekilde da\u011f\u0131ld\u0131\u011f\u0131n\u0131<\/a> varsayar.<\/span><\/p>\n Bu hipotezin do\u011fru olup olmad\u0131\u011f\u0131n\u0131 kontrol etmenin iki yolu vard\u0131r:<\/span><\/p>\n 1. Histogram veya QQ grafi\u011fi olu\u015fturun<\/strong><\/span><\/p>\n Bir histogram veya QQ grafi\u011fi olu\u015fturarak yan\u0131t de\u011fi\u015fkeninin da\u011f\u0131l\u0131m\u0131n\u0131n yakla\u015f\u0131k olarak normal da\u011f\u0131l\u0131p da\u011f\u0131lmad\u0131\u011f\u0131n\u0131 g\u00f6rsel olarak kontrol edebilirsiniz.<\/span><\/p>\n Bir histogram<\/strong> olu\u015fturursan\u0131z, yan\u0131t de\u011fi\u015fkeninin da\u011f\u0131l\u0131m\u0131n\u0131n yakla\u015f\u0131k olarak bir “\u00e7an” \u015feklini takip edip etmedi\u011fini kontrol etmeniz yeterlidir. E\u011fer \u00f6yleyse, genellikle normallik varsay\u0131m\u0131n\u0131n kar\u015f\u0131land\u0131\u011f\u0131n\u0131 varsayabilirsiniz:<\/span> <\/p>\n Bir QQ grafi\u011fi<\/strong> olu\u015fturuyorsan\u0131z, veri noktalar\u0131n\u0131n d\u00fcz bir \u00e7apraz \u00e7izgi \u00fczerinde olup olmad\u0131\u011f\u0131n\u0131 kontrol edin. E\u011fer \u00f6yleyse, genel olarak normallik varsay\u0131m\u0131n\u0131n kar\u015f\u0131land\u0131\u011f\u0131n\u0131 varsayabilirsiniz:<\/span> <\/p>\n \u0130lgili:<\/strong><\/span> Normalli\u011fi Kontrol Etmek \u0130\u00e7in QQ Grafikleri Nas\u0131l Kullan\u0131l\u0131r?<\/a><\/p>\n 2. Resmi bir istatistiksel test yap\u0131n<\/strong><\/span><\/p>\n Normalli\u011fi kontrol etmek i\u00e7in Shapiro-Wilk testi<\/a> de yapabilirsiniz. Testin p de\u011ferinin<\/a> 0,05’ten k\u00fc\u00e7\u00fck olmas\u0131 verilerin normal da\u011f\u0131lmad\u0131\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n Ancak \u00e7ok b\u00fcy\u00fck \u00f6rneklerle \u00e7al\u0131\u015f\u0131rken Shapiro-Wilk testi gibi istatistiksel testlerin neredeyse her zaman verilerinizin normal olmad\u0131\u011f\u0131n\u0131 s\u00f6yleyece\u011fini unutmay\u0131n.<\/span><\/p>\n Bu nedenle, histogramlar ve QQ grafikleri gibi grafikleri kullanarak verilerinizi g\u00f6rsel olarak incelemek genellikle en iyisidir. Sadece grafiklere bakarak verilerin normal da\u011f\u0131l\u0131p da\u011f\u0131lmad\u0131\u011f\u0131na dair olduk\u00e7a iyi bir fikir edinebilirsiniz.<\/span><\/p>\n Genel olarak, tekrarlanan \u00f6l\u00e7\u00fcm ANOVA’s\u0131n\u0131n, \u00f6rneklem boyutlar\u0131 yeterince b\u00fcy\u00fck oldu\u011fu s\u00fcrece normallik varsay\u0131m\u0131n\u0131n ihlallerine kar\u015f\u0131 olduk\u00e7a sa\u011flam oldu\u011fu kabul edilir.<\/span><\/p>\n Normallik varsay\u0131m\u0131 ciddi \u015fekilde ihlal edilirse iki se\u00e7ene\u011finiz vard\u0131r:<\/span><\/p>\n 1.<\/strong> Da\u011f\u0131l\u0131mlar\u0131n daha normal da\u011f\u0131lmas\u0131 i\u00e7in verilerinizin yan\u0131t de\u011ferlerini d\u00f6n\u00fc\u015ft\u00fcr\u00fcn<\/a> .<\/span><\/p>\n 2.<\/strong> Friedman testi<\/a> gibi normallik varsay\u0131m\u0131n\u0131 gerektirmeyen e\u015fde\u011fer bir parametrik olmayan test ger\u00e7ekle\u015ftirin.<\/span><\/p>\n Tekrarlanan \u00f6l\u00e7\u00fcmlerde ANOVA k\u00fcreselli\u011fi<\/strong> varsayar; yani ilgili gruplar\u0131n t\u00fcm kombinasyonlar\u0131 aras\u0131ndaki farklar\u0131n varyanslar\u0131 e\u015fit olmal\u0131d\u0131r.<\/span><\/p>\n Bu varsay\u0131m kar\u015f\u0131lanmazsa F oran\u0131 \u015fi\u015firilir ve tekrarlanan \u00f6l\u00e7\u00fcmlerin sonu\u00e7lar\u0131 ANOVA g\u00fcvenilmez hale gelir.<\/span><\/p>\n Bu varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131n\u0131 test etmek i\u00e7in Mauchly’nin k\u00fcresellik testini ger\u00e7ekle\u015ftirebiliriz.<\/span><\/p>\n Bu test a\u015fa\u011f\u0131daki bo\u015f ve alternatif hipotezi kullan\u0131r:<\/span><\/p>\n Testin p de\u011feri belirli bir anlaml\u0131l\u0131k d\u00fczeyinin<\/a> alt\u0131ndaysa (\u00f6rne\u011fin \u03b1 = 0,05), o zaman s\u0131f\u0131r hipotezini reddederiz ve farklar\u0131n varyanslar\u0131n\u0131n e\u015fit olmad\u0131\u011f\u0131 sonucuna var\u0131r\u0131z.<\/span><\/p>\n Aksi halde, e\u011fer p-de\u011feri belirli bir anlaml\u0131l\u0131k seviyesinden az de\u011filse (\u00f6rn. \u03b1 = 0.05), s\u0131f\u0131r hipotezini reddetmede ba\u015far\u0131s\u0131z oluruz ve k\u00fcresellik varsay\u0131m\u0131n\u0131n kar\u015f\u0131land\u0131\u011f\u0131 sonucuna var\u0131r\u0131z.<\/span><\/p>\n Kulland\u0131\u011f\u0131n\u0131z istatistiksel yaz\u0131l\u0131ma ba\u011fl\u0131 olarak bu testin sonu\u00e7lar\u0131 \u015f\u00f6yle g\u00f6r\u00fcnecektir:<\/span> <\/p>\n P de\u011feri 0,05’ten az olmad\u0131\u011f\u0131 i\u00e7in s\u0131f\u0131r hipotezini reddetmede ba\u015far\u0131s\u0131z olaca\u011f\u0131z ve bu \u00f6zel \u00f6rnekte k\u00fcresellik varsay\u0131m\u0131n\u0131n kar\u015f\u0131land\u0131\u011f\u0131 sonucuna varaca\u011f\u0131z.<\/span><\/p>\n Mauchly k\u00fcresellik testinin bo\u015f hipotezini reddedersek, tekrarlanan \u00f6l\u00e7\u00fcmler ANOVA tablosundaki F de\u011ferini hesaplamak i\u00e7in kullan\u0131lan serbestlik derecelerine genellikle bir d\u00fczeltme uygular\u0131z.<\/span><\/p>\n Uygulayabilece\u011fimiz \u00fc\u00e7 d\u00fczeltme var:<\/span><\/p>\n Bu d\u00fczeltmelerin her biri, k\u00fcresellik varsay\u0131m\u0131n\u0131n ihlal edildi\u011fi ger\u00e7e\u011fini hesaba katmak i\u00e7in tekrarlanan \u00f6l\u00e7\u00fcmler ANOVA \u00e7\u0131kt\u0131 tablosundaki p de\u011ferlerini art\u0131rma e\u011filimindedir.<\/span><\/p>\n Daha sonra bu p de\u011ferlerini, tekrarlanan ANOVA \u00f6l\u00e7\u00fcmlerinin s\u0131f\u0131r hipotezini reddedip reddetmememiz gerekti\u011fini belirlemek i\u00e7in kullanabiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki e\u011fitimler tekrarlanan \u00f6l\u00e7\u00fcmler ANOVA hakk\u0131nda ek bilgi sa\u011flar:<\/span><\/p>\n Tekrarlanan \u00d6l\u00e7\u00fcmlere Giri\u015f ANOVA<\/a> Tekrarlanan \u00f6l\u00e7\u00fcmler ANOVA, her grupta ayn\u0131 deneklerin yer ald\u0131\u011f\u0131 \u00fc\u00e7 veya daha fazla grubun ortalamalar\u0131 aras\u0131nda istatistiksel olarak anlaml\u0131 bir fark olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in kullan\u0131l\u0131r. Ancak tekrarlanan \u00f6l\u00e7\u00fcm ANOVA’s\u0131n\u0131 ger\u00e7ekle\u015ftirmeden \u00f6nce a\u015fa\u011f\u0131daki varsay\u0131mlar\u0131n kar\u015f\u0131land\u0131\u011f\u0131ndan emin olmal\u0131y\u0131z: 1. Ba\u011f\u0131ms\u0131zl\u0131k: Her g\u00f6zlem ba\u011f\u0131ms\u0131z olmal\u0131d\u0131r. 2. Normallik: Yan\u0131t de\u011fi\u015fkeninin da\u011f\u0131l\u0131m\u0131 normal da\u011f\u0131lm\u0131\u015ft\u0131r. 3. K\u00fcresellik: \u0130lgili gruplar\u0131n t\u00fcm […]<\/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-2462","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n Hipotez 1: Ba\u011f\u0131ms\u0131zl\u0131k<\/strong><\/span><\/h2>\n
Bu varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131 nas\u0131l belirlenir<\/strong><\/h3>\n
Bu varsay\u0131ma uyulmazsa ne yap\u0131lmal\u0131?<\/strong><\/h3>\n
Hipotez 2: normallik<\/strong><\/span><\/h2>\n
Bu varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131 nas\u0131l belirlenir<\/strong><\/h3>\n
![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/representant11.png\")
<\/p>\n![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/rep12.png\")
<\/p>\n Bu varsay\u0131ma uyulmazsa ne yap\u0131lmal\u0131?<\/strong><\/h3>\n
Varsay\u0131m 3: K\u00fcresellik<\/strong><\/span><\/h2>\n
Bu varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131 nas\u0131l belirlenir<\/strong><\/h3>\n
\n
![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/mauchly2.png\")
<\/p>\n Bu varsay\u0131ma uyulmazsa ne yap\u0131lmal\u0131?<\/strong><\/h3>\n
\n
Ek kaynaklar<\/strong><\/span><\/h3>\n
Tekrarlanan \u00d6l\u00e7\u00fcmler ANOVA Hesaplay\u0131c\u0131<\/a>
Tekrarlanan \u00f6l\u00e7\u00fcmlerin sonu\u00e7lar\u0131 ANOVA nas\u0131l raporlan\u0131r?<\/a>
Tek y\u00f6nl\u00fc ANOVA ve tekrarlanan \u00f6l\u00e7\u00fcmler ANOVA: fark<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"