{"id":104,"date":"2008-08-20T18:07:44","date_gmt":"2008-08-20T15:07:44","guid":{"rendered":"http:\/\/www.muratakyildiz.com\/wordpress\/?p=104"},"modified":"2025-03-28T23:27:34","modified_gmt":"2025-03-28T20:27:34","slug":"negatif-guevenirlik-katsays-ve-olas-sebepleri","status":"publish","type":"post","link":"https:\/\/www.istatistik.gen.tr\/?p=104","title":{"rendered":"Negatif g\u00fcvenirlik katsay\u0131s\u0131 ve olas\u0131 sebepleri"},"content":{"rendered":"<div align=\"justify\">\u00c7o\u011funlukla bu duruma negatif madde kovaryanslar\u0131 yol \u00e7ar. Yani maddelerin ayn\u0131 de\u011fi\u015fkene \u00f6l\u00e7meye y\u00f6nelmi\u015f&nbsp;olmamalar\u0131 ve i\u00e7 tutarl\u0131l\u0131\u011f\u0131n son derece bozuk oldu\u011fu durumlar. \u00d6rne\u011fin ilk bir ka\u00e7 maddeye verilen tepkilerin \u00f6r\u00fcnt\u00fcs\u00fcyle sondaki maddelere verilen tepkilerin \u00f6r\u00fcnt\u00fcs\u00fcn birbirini tutmad\u0131\u011f\u0131 durumlar.<\/div>\n<div align=\"justify\"><\/div>\n<p><!--more--><\/p>\n<div align=\"justify\">Temelde, i\u00e7 tutarl\u0131l\u0131k anlam\u0131nda g\u00fcvenilirlik, t\u00fcm maddelerin ayn\u0131 \u015feyi \u00f6l\u00e7t\u00fcklerini&nbsp; varsayar. Dolay\u0131s\u0131 ile de i\u00e7 tutarl\u0131l\u0131k anlam\u0131nda g\u00fcvenilirli\u011fin y\u00fckselmesi \u00f6l\u00e7\u00fclen kapsam\u0131n daralmas\u0131 anlam\u0131na gelir. Bu nedenle i\u00e7i tutarl\u0131l\u0131\u011f\u0131 bildirmesi i\u00e7in hesaplanan g\u00fcvenilirlik katsay\u0131lar\u0131n\u0131n d\u00fc\u015f\u00fck \u00e7\u0131kmas\u0131, \u00f6l\u00e7e\u011fin birden \u00e7ok ve birbirinden ba\u011f\u0131ms\u0131z fakt\u00f6re sahip olabilece\u011fini de akla getirmelidir. Bu durumda, maddeler aras\u0131 kovaryans\u0131n negatif olmas\u0131na benzer olarak, \u00f6l\u00e7e\u011fin i\u00e7inde yer alan fakt\u00f6rlerin birbirleriyle s\u0131f\u0131r&#8217;a yak\u0131n ya da negatif korelasyon g\u00f6stermeleri de negatif g\u00fcvenilirlik katsay\u0131lar\u0131n\u0131n bulunmas\u0131na yol a\u00e7abilir.Form\u00fclden hat\u0131rlanaca\u011f\u0131 gibi, KR20&#8217;ye dayal\u0131 olarak negatif bir katsay\u0131 hesaplayabilmenin tek yolu, madde varyanslar\u0131 toplam\u0131n\u0131n test varyans\u0131 toplam\u0131ndan daha b\u00fcy\u00fck olmas\u0131d\u0131r. Bunu ger\u00e7ekle\u015ftirebilecek en iyi durum maddelerin birbirinden ba\u011f\u0131ms\u0131z \u00f6zellikleri \u00f6l\u00e7mesidir. B\u00f6ylece madde varyanslar\u0131 b\u00fcy\u00fcyecektir.<\/p>\n<p>Ayn\u0131 \u015feyi simgelerle g\u00f6sterirsek:<\/p>\n<\/div>\n<div align=\"justify\">\u03a3p x q &gt; \u03a3X<sup>2<\/sup><\/div>\n<div align=\"justify\"><\/div>\n<div align=\"justify\">yani madde puanlar\u0131 varyans\u0131 test puanlar\u0131 varyans\u0131ndan b\u00fcy\u00fck olmal\u0131d\u0131r\u03a3X<sup>2<\/sup> = \u03a3p*q + 2*\u03a3Covij<\/p>\n<\/div>\n<div align=\"justify\"><\/div>\n<div align=\"justify\">yani test puanlar\u0131 varyans\u0131, madde puanlar\u0131 varyans\u0131 + madde kovaryanslar\u0131<br \/>\ntoplam\u0131n\u0131n iki kat\u0131na e\u015fittir.Madde puanlar\u0131 varyans\u0131n\u0131n test puanlar\u0131 varyans\u0131ndan b\u00fcy\u00fck olabilmesi i\u00e7in 2*\u03a3Covij de\u011ferinin mutlaka negatif olmas\u0131 gerekir. 2*\u03a3Covij de\u011feri, maddeler aras\u0131 ortalama kovaryans olarak adland\u0131r\u0131l\u0131r. Dolay\u0131s\u0131 ile ters y\u00f6nde d\u00fc\u015f\u00fcn\u00fcrsek y\u00fcksek i\u00e7 tutarl\u0131l\u0131k anlam\u0131nda g\u00fcvenilirli\u011fi elde edebilmenin yolu bu de\u011feri y\u00fckseltmektir. Yani maddeler aras\u0131ndaki ortak varyans\u0131 artt\u0131rmakt\u0131r.<\/p>\n<p>Benzer durumun birbirinden olduk\u00e7a ba\u011f\u0131ms\u0131z ya da aralar\u0131nda negatif korelasyon bulunan fakt\u00f6rlerden olu\u015fan \u00f6l\u00e7eklerde de olu\u015fabildi\u011fini s\u00f6yleyebiliriz. Bunun sebebi ise yine birbirinden ba\u011f\u0131ms\u0131z alt\u00f6l\u00e7eklerde yer alan maddeler aras\u0131ndaki ortak varyans\u0131n \u00e7ok d\u00fc\u015f\u00fck ya da negatif olacak olmas\u0131d\u0131r. Y\u00fcksek \u00f6zde\u011fere (eigenvalue) sahip (dolay\u0131s\u0131 ile kendi maddeleri aras\u0131ndaki korelasyonlar\u0131 y\u00fcksek olan) birbirinden olduk\u00e7a b\u011f\u0131ms\u0131z ya da birbirleri aras\u0131nda negatif korelasyon bulunan fakt\u00f6rlerin olu\u015fturaca\u011f\u0131 iki \u00f6l\u00e7\u00fcm\u00fcn olu\u015fturaca\u011f\u0131 b\u00fct\u00fcn\u00fcn sunaca\u011f\u0131 korelasyon katsay\u0131s\u0131 \u00e7o\u011funlukla s\u0131f\u0131ra ya da negatif bir korelasyon katsay\u0131s\u0131na e\u015fit olur.<\/p>\n<\/div>\n<div align=\"justify\"><\/div>\n<div align=\"justify\">Ayr\u0131ca ek olarak da\u011f\u0131l\u0131mlar\u0131n a\u015f\u0131r\u0131 \u00e7arp\u0131k oldu\u011fu durumlarda, korelasyon katsay\u0131s\u0131n\u0131n negatif olmas\u0131 yetmiyormu\u015f gibi -1.65 ya da -2 gibi korelasyon katsay\u0131lar\u0131 bile hesaplanabildi\u011fi bildirilmi\u015ftir<\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u00c7o\u011funlukla bu duruma negatif madde kovaryanslar\u0131 yol \u00e7ar. Yani maddelerin ayn\u0131 de\u011fi\u015fkene \u00f6l\u00e7meye y\u00f6nelmi\u015f&nbsp;olmamalar\u0131 ve i\u00e7 tutarl\u0131l\u0131\u011f\u0131n son derece bozuk oldu\u011fu durumlar. \u00d6rne\u011fin ilk bir ka\u00e7 maddeye verilen tepkilerin \u00f6r\u00fcnt\u00fcs\u00fcyle sondaki maddelere verilen tepkilerin \u00f6r\u00fcnt\u00fcs\u00fcn birbirini tutmad\u0131\u011f\u0131 durumlar.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":true,"template":"","format":"standard","meta":{"_kad_post_transparent":"","_kad_post_title":"","_kad_post_layout":"","_kad_post_sidebar_id":"","_kad_post_content_style":"","_kad_post_vertical_padding":"","_kad_post_feature":"","_kad_post_feature_position":"","_kad_post_header":false,"_kad_post_footer":false,"_kad_post_classname":"","footnotes":""},"categories":[1],"tags":[],"class_list":["post-104","post","type-post","status-publish","format-standard","hentry","category-genel"],"_links":{"self":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts\/104","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=104"}],"version-history":[{"count":3,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts\/104\/revisions"}],"predecessor-version":[{"id":443,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts\/104\/revisions\/443"}],"wp:attachment":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=104"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=104"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=104"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}