\u00d6rne\u011fin, e\u011fitim seviyesinin (lise diplomas\u0131, \u00f6nlisans, lisans, y\u00fcksek lisans) hem y\u0131ll\u0131k geliri hem de toplam \u00f6\u011frenci kredisi borcunu nas\u0131l etkiledi\u011fini analiz etmek i\u00e7in MANOVA’y\u0131 kullanabiliriz.<\/span> <\/p>\n \u0130lgili:<\/strong><\/span> ANOVA, ANCOVA, MANOVA ve MANCOVA aras\u0131ndaki farklar<\/a><\/p>\n Her MANOVA ger\u00e7ekle\u015ftirdi\u011fimizde a\u015fa\u011f\u0131daki varsay\u0131mlar\u0131n kar\u015f\u0131land\u0131\u011f\u0131n\u0131 do\u011frulamam\u0131z gerekir:<\/span><\/p>\n 1. \u00c7ok de\u011fi\u015fkenli normallik<\/strong> \u2013 Yan\u0131t de\u011fi\u015fkenleri, her fakt\u00f6r de\u011fi\u015fken(ler)i grubu i\u00e7inde \u00e7ok de\u011fi\u015fkenli normal olarak da\u011f\u0131t\u0131l\u0131r.<\/span><\/p>\n 2. Ba\u011f\u0131ms\u0131zl\u0131k<\/strong> \u2013 Her g\u00f6zlem pop\u00fclasyondan rastgele ve ba\u011f\u0131ms\u0131z olarak \u00f6rneklenir.<\/span><\/p>\n 3. E\u015fit varyans<\/strong> \u2013 Her grubun pop\u00fclasyon kovaryans matrisleri e\u015fittir.<\/span><\/p>\n 4. \u00c7ok de\u011fi\u015fkenli ayk\u0131r\u0131 de\u011ferler yok<\/strong> \u2013 A\u015f\u0131r\u0131 \u00e7ok de\u011fi\u015fkenli ayk\u0131r\u0131 de\u011ferler yoktur.<\/span><\/p>\n Bu makalede, her bir varsay\u0131m\u0131n a\u00e7\u0131klanmas\u0131n\u0131n yan\u0131 s\u0131ra, varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131n\u0131n nas\u0131l belirlenece\u011fi de a\u00e7\u0131klanmaktad\u0131r.<\/span><\/p>\n MANOVA, yan\u0131t de\u011fi\u015fkenlerinin, fakt\u00f6r de\u011fi\u015fkeninin her grubu i\u00e7inde \u00e7ok de\u011fi\u015fkenli, normal olarak da\u011f\u0131ld\u0131\u011f\u0131n\u0131 varsayar.<\/span><\/p>\n Her fakt\u00f6r * yan\u0131t de\u011fi\u015fkeni kombinasyonu i\u00e7in en az 20 g\u00f6zlem varsa, \u00e7ok de\u011fi\u015fkenli normallik varsay\u0131m\u0131n\u0131n kar\u015f\u0131land\u0131\u011f\u0131n\u0131 varsayabiliriz.<\/span><\/p>\n Her fakt\u00f6r*yan\u0131t de\u011fi\u015fkeni kombinasyonu i\u00e7in 20’den az g\u00f6zlem varsa, art\u0131klar\u0131 g\u00f6rselle\u015ftirmek ve bu varsay\u0131m\u0131n kar\u015f\u0131lan\u0131p kar\u015f\u0131lanmad\u0131\u011f\u0131n\u0131 g\u00f6rsel olarak kontrol etmek i\u00e7in bir da\u011f\u0131l\u0131m grafi\u011fi matrisi olu\u015fturabiliriz.<\/span><\/p>\n Neyse ki MANOVA’n\u0131n \u00e7ok de\u011fi\u015fkenli normallikten sapmalara kar\u015f\u0131 dayan\u0131kl\u0131 oldu\u011fu iyi bilinmektedir, bu nedenle k\u00fc\u00e7\u00fck ila orta dereceli sapmalar genellikle sorun de\u011fildir.<\/span><\/p>\n MANOVA, her g\u00f6zlemin pop\u00fclasyondan rastgele ve ba\u011f\u0131ms\u0131z olarak \u00f6rneklendi\u011fini varsayar.<\/span><\/p>\n Verileri toplamak i\u00e7in olas\u0131l\u0131kl\u0131 \u00f6rnekleme y\u00f6ntemi<\/a> (bir pop\u00fclasyonun her \u00fcyesinin \u00f6rneklemde se\u00e7ilme olas\u0131l\u0131\u011f\u0131 e\u015fittir) kullan\u0131ld\u0131\u011f\u0131 s\u00fcrece, her g\u00f6zlemin rastgele ve ba\u011f\u0131ms\u0131z bir \u015fekilde \u00f6rneklendi\u011fini varsayabiliriz.<\/span><\/p>\n Olas\u0131l\u0131k \u00f6rnekleme y\u00f6ntemlerinin \u00f6rnekleri \u015funlar\u0131 i\u00e7erir:<\/span><\/p>\n MANOVA, her grubun pop\u00fclasyon kovaryans matrislerinin e\u015fit oldu\u011funu varsayar.<\/span><\/p>\n Bu hipotezi test etmenin en yayg\u0131n yolu Box’\u0131n M testini kullanmakt\u0131r. Bu testin olduk\u00e7a kat\u0131 oldu\u011fu bilinmektedir, bu nedenle pop\u00fclasyon kovaryans matrislerinin e\u015fit olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in genellikle 0,001 anlaml\u0131l\u0131k d\u00fczeyini kullan\u0131r\u0131z.<\/span><\/p>\n Box’\u0131n M testinin p de\u011feri 0,001’den b\u00fcy\u00fckse bu varsay\u0131m\u0131n kar\u015f\u0131land\u0131\u011f\u0131n\u0131 varsayabiliriz.<\/span><\/p>\n Neyse ki testin p de\u011feri 0,001’den k\u00fc\u00e7\u00fck olsa bile MANOVA bu hipotezden sapmalara kar\u015f\u0131 dayan\u0131kl\u0131 olma e\u011filimindedir.<\/span><\/p>\n E\u015fit olmayan kovaryans matrislerinin sorun olabilmesi i\u00e7in kovaryans matrisleri aras\u0131ndaki farklar\u0131n olduk\u00e7a a\u015f\u0131r\u0131 olmas\u0131 gerekir.<\/span><\/p>\n MANOVA, verilerde sonu\u00e7lar\u0131 \u00f6nemli \u00f6l\u00e7\u00fcde etkileyebilecek a\u015f\u0131r\u0131 \u00e7ok de\u011fi\u015fkenli ayk\u0131r\u0131 de\u011ferlerin bulunmad\u0131\u011f\u0131n\u0131 varsayar.<\/span><\/p>\n Bu varsay\u0131m\u0131 do\u011frulaman\u0131n en yayg\u0131n yolu, her g\u00f6zlem i\u00e7in \u00e7ok de\u011fi\u015fkenli bir uzayda iki nokta aras\u0131ndaki mesafeyi temsil eden Mahalanobis mesafesini hesaplamakt\u0131r.<\/span><\/p>\n Bir g\u00f6zlemin Mahalanobis mesafesine kar\u015f\u0131l\u0131k gelen p de\u011feri 0,001’den k\u00fc\u00e7\u00fckse, genellikle bu g\u00f6zlemin a\u015f\u0131r\u0131 u\u00e7 de\u011fer oldu\u011funu beyan ederiz.<\/span><\/p>\n \u00c7e\u015fitli istatistiksel yaz\u0131l\u0131mlarda Mahalanobis mesafesinin nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6rmek i\u00e7in a\u015fa\u011f\u0131daki e\u011fitimlere bak\u0131n:<\/span><\/p>\n A\u015fa\u011f\u0131daki e\u011fitimlerde \u00e7e\u015fitli istatistiksel yaz\u0131l\u0131mlarda MANOVA’n\u0131n nas\u0131l ger\u00e7ekle\u015ftirilece\u011fi a\u00e7\u0131klanmaktad\u0131r:<\/span><\/p>\n R’de MANOVA nas\u0131l ger\u00e7ekle\u015ftirilir<\/a> MANOVA (\u00e7ok de\u011fi\u015fkenli varyans analizi), bir veya daha fazla fakt\u00f6r de\u011fi\u015fkeninin birden fazla yan\u0131t de\u011fi\u015fkenini nas\u0131l etkiledi\u011fini analiz etmek i\u00e7in kullan\u0131l\u0131r. \u00d6rne\u011fin, e\u011fitim seviyesinin (lise diplomas\u0131, \u00f6nlisans, lisans, y\u00fcksek lisans) hem y\u0131ll\u0131k geliri hem de toplam \u00f6\u011frenci kredisi borcunu nas\u0131l etkiledi\u011fini analiz etmek i\u00e7in MANOVA’y\u0131 kullanabiliriz. \u0130lgili: ANOVA, ANCOVA, MANOVA ve MANCOVA aras\u0131ndaki farklar Her […]<\/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-2015","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/manova.png\")
<\/p>\n Varsay\u0131m 1: \u00c7ok de\u011fi\u015fkenli normallik<\/strong><\/span><\/h3>\n
Hipotez 2: Ba\u011f\u0131ms\u0131zl\u0131k<\/strong><\/span><\/h3>\n
\n
Varsay\u0131m 3: e\u015fit varyans<\/strong><\/span><\/h3>\n
Hipotez 4: \u00e7ok de\u011fi\u015fkenli ayk\u0131r\u0131 de\u011ferler yok<\/strong><\/span><\/h3>\n
\n
Ek kaynaklar<\/strong><\/span><\/h3>\n
SPSS’de MANOVA Nas\u0131l Ger\u00e7ekle\u015ftirilir<\/a>
Stata’da MANOVA nas\u0131l ger\u00e7ekle\u015ftirilir?<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"