{"id":1301,"date":"2023-07-26T23:09:57","date_gmt":"2023-07-26T23:09:57","guid":{"rendered":"https:\/\/statorials.org\/tr\/tukey-test-pitonu\/"},"modified":"2023-07-26T23:09:57","modified_gmt":"2023-07-26T23:09:57","slug":"tukey-test-pitonu","status":"publish","type":"post","link":"https:\/\/statorials.org\/tr\/tukey-test-pitonu\/","title":{"rendered":"Python'da tukey testi nas\u0131l yap\u0131l\u0131r?"},"content":{"rendered":"

<\/p>\n

\n

\u00dc\u00e7 veya daha fazla ba\u011f\u0131ms\u0131z grubun ortalamalar\u0131 aras\u0131nda istatistiksel olarak anlaml\u0131 bir fark olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in tek y\u00f6nl\u00fc ANOVA<\/a> kullan\u0131l\u0131r.<\/span><\/p>\n

ANOVA tablosunun genel p de\u011feri belirli bir anlaml\u0131l\u0131k d\u00fczeyinin alt\u0131ndaysa, grup ortalamalar\u0131ndan en az birinin di\u011ferlerinden farkl\u0131 oldu\u011funu s\u00f6ylemek i\u00e7in yeterli kan\u0131t\u0131m\u0131z var demektir.<\/span><\/p>\n

Ancak bu bize hangi<\/em> gruplar\u0131n birbirinden farkl\u0131 oldu\u011funu s\u00f6ylemez. Bu bize basit\u00e7e t\u00fcm grup ortalamalar\u0131n\u0131n e\u015fit olmad\u0131\u011f\u0131n\u0131 s\u00f6yler. Hangi gruplar\u0131n birbirinden farkl\u0131 oldu\u011funu tam olarak bilmek i\u00e7in post hoc test<\/a> yapmam\u0131z gerekir.<\/span><\/p>\n

En s\u0131k kullan\u0131lan post hoc testlerden biri, aile baz\u0131nda hata oran\u0131n\u0131<\/a> kontrol ederken her grubun ortalamalar\u0131 aras\u0131nda ikili kar\u015f\u0131la\u015ft\u0131rmalar yapmam\u0131za olanak tan\u0131yan Tukey testidir<\/strong> .<\/span><\/p>\n

Bu e\u011fitimde Python’da Tukey testinin nas\u0131l ger\u00e7ekle\u015ftirilece\u011fine ili\u015fkin ad\u0131m ad\u0131m bir \u00f6rnek sunulmaktad\u0131r.<\/span><\/p>\n

1. Ad\u0131m: Gerekli paketleri ve i\u015flevleri y\u00fckleyin<\/strong><\/span><\/h3>\n

\u00d6ncelikle gerekli paketleri ve fonksiyonlar\u0131 Python’a y\u00fckleyece\u011fiz:<\/span><\/p>\n

 import<\/span> pandas as<\/span> pd\nimport<\/span> numpy as<\/span> np\nfrom<\/span> scipy. stats<\/span> import<\/span> f_oneway\nfrom<\/span> statsmodels. stats<\/span> . multicomp<\/span> import<\/span> pairwise_tukeyhsd<\/strong><\/span><\/pre>\n
 Ad\u0131m 2: ANOVA modelini yerle\u015ftirin<\/strong><\/span><\/h3>\n
 A\u015fa\u011f\u0131daki kod, \u00fc\u00e7 grupla (A, B ve C) sahte bir veri k\u00fcmesinin nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 ve her grubun ortalama de\u011ferlerinin e\u015fit olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in verilere tek y\u00f6nl\u00fc bir ANOVA modelinin nas\u0131l yerle\u015ftirilece\u011fini g\u00f6sterir:<\/span><\/p>\n
 #enter data for three groups\n<\/span>a = [85, 86, 88, 75, 78, 94, 98, 79, 71, 80]\nb = [91, 92, 93, 90, 97, 94, 82, 88, 95, 96]\nc = [79, 78, 88, 94, 92, 85, 83, 85, 82, 81]\n\n#perform one-way ANOVA\n<\/span>f_oneway(a, b, c)\n\nF_onewayResult(statistic=5.167774552944481, pvalue=0.012582197136592609)<\/strong><\/pre>\n
 ANOVA tablosundan genel p de\u011ferinin 0,01258<\/strong> oldu\u011funu g\u00f6rebiliriz.<\/span><\/p>\n
 Bu say\u0131 0,05’ten k\u00fc\u00e7\u00fck oldu\u011fundan her gruptaki ortalama de\u011ferlerin e\u015fit olmad\u0131\u011f\u0131n\u0131 s\u00f6yleyebilecek yeterli kan\u0131t\u0131m\u0131z var.<\/span><\/p>\n
 Yani tam olarak hangi grup ortalamalar\u0131n\u0131n farkl\u0131 oldu\u011funu belirlemek i\u00e7in Tukey testini yapabiliriz.<\/span><\/p>\n
 3. Ad\u0131m: Tukey testini ger\u00e7ekle\u015ftirin<\/strong><\/span><\/h3>\n
 Python’da Tukey testini ger\u00e7ekle\u015ftirmek i\u00e7in statsmodels<\/strong> k\u00fct\u00fcphanesindekipairwise_tukeyhsd ()<\/strong> fonksiyonunu kullanabiliriz:<\/span><\/p>\n
 #create DataFrame to hold data\n<\/span>df = pd. DataFrame<\/span> ({'score': [85, 86, 88, 75, 78, 94, 98, 79, 71, 80,\n                             91, 92, 93, 90, 97, 94, 82, 88, 95, 96,\n                             79, 78, 88, 94, 92, 85, 83, 85, 82, 81],\n                   'group': np. repeat<\/span> (['a', 'b', 'c'], repeats= 10<\/span> )}) \n\n# perform Tukey's test\n<\/span>tukey = pairwise_tukeyhsd(endog=df['score'],\n                          groups=df['group'],\n                          alpha= 0.05<\/span> )\n\n#display results\n<\/span>print<\/span> (tukey)\n\n Multiple Comparison of Means - Tukey HSD, FWER=0.05 \n==================================================== ===\ngroup1 group2 meandiff p-adj lower upper reject\n-------------------------------------------------- ---\n     ab 8.4 0.0158 1.4272 15.3728 True\n     ac 1.3 0.8864 -5.6728 8.2728 False\n     bc -7.1 0.0453 -14.0728 -0.1272 True\n-------------------------------------------------- ---<\/strong><\/pre>\n
 Sonucun nas\u0131l yorumlanaca\u011f\u0131 a\u015fa\u011f\u0131da a\u00e7\u0131klanm\u0131\u015ft\u0131r:<\/span><\/span><\/p>\n