<\/p>\n
A\u015fa\u011f\u0131daki \u00f6rneklerde pandas DataFrame kullan\u0131larak \u00fc\u00e7 farkl\u0131 t testinin nas\u0131l ger\u00e7ekle\u015ftirilece\u011fi g\u00f6sterilmektedir:<\/span><\/p>\n \u0130ki pop\u00fclasyonun ortalamalar\u0131n\u0131n e\u015fit olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in ba\u011f\u0131ms\u0131z iki \u00f6rnekli t testi<\/a> kullan\u0131l\u0131r.<\/span><\/p>\n \u00d6rne\u011fin, bir profes\u00f6r\u00fcn iki farkl\u0131 \u00e7al\u0131\u015fma y\u00f6nteminin s\u0131navlarda farkl\u0131 ortalama notlara yol a\u00e7\u0131p a\u00e7mad\u0131\u011f\u0131n\u0131 bilmek istedi\u011fini varsayal\u0131m.<\/span><\/p>\n Bunu test etmek i\u00e7in A y\u00f6ntemini kullanacak 10 \u00f6\u011frenciyi ve B y\u00f6ntemini kullanacak 10 \u00f6\u011frenciyi i\u015fe al\u0131r.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, her \u00f6\u011frencinin puanlar\u0131n\u0131n bir pandas DataFrame’e nas\u0131l girilece\u011fini ve ard\u0131ndan ba\u011f\u0131ms\u0131z iki \u00f6rnekli bir t testi ger\u00e7ekle\u015ftirmek i\u00e7in SciPy<\/strong> k\u00fct\u00fcphanesinin ttest_ind()<\/a> i\u015flevinin nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Sonu\u00e7tan \u015funu g\u00f6rebiliriz:<\/span><\/span><\/p>\n P de\u011feri 0,05’ten k\u00fc\u00e7\u00fck oldu\u011fundan, t testinin bo\u015f hipotezini reddediyoruz ve iki y\u00f6ntemin farkl\u0131 ortalama s\u0131nav puanlar\u0131na yol a\u00e7t\u0131\u011f\u0131n\u0131 belirtmek i\u00e7in yeterli kan\u0131t oldu\u011fu sonucuna var\u0131yoruz.<\/span><\/p>\n Welch’in t testi,<\/a> ba\u011f\u0131ms\u0131z iki \u00f6rnekli t testine benzer, tek fark\u0131,<\/a> \u00f6rneklerin al\u0131nd\u0131\u011f\u0131 iki pop\u00fclasyonun e\u015fit varyansa sahip oldu\u011funu varsaymamas\u0131d\u0131r.<\/span><\/p>\n Welch’in t-testini \u00f6nceki \u00f6rnekle tamamen ayn\u0131 veri k\u00fcmesi \u00fczerinde ger\u00e7ekle\u015ftirmek i\u00e7in, ttest_ind<\/strong> () i\u015flevinde equal_var=False<\/strong> de\u011ferini a\u015fa\u011f\u0131daki gibi belirtmemiz yeterlidir:<\/span><\/p>\n Sonu\u00e7tan \u015funu g\u00f6rebiliriz:<\/span><\/p>\n P de\u011feri 0,05’ten k\u00fc\u00e7\u00fck oldu\u011fundan Welch t testinin bo\u015f hipotezini reddediyoruz ve iki y\u00f6ntemin farkl\u0131 ortalama s\u0131nav puanlar\u0131na yol a\u00e7t\u0131\u011f\u0131n\u0131 ifade etmek i\u00e7in yeterli kan\u0131t oldu\u011fu sonucuna var\u0131yoruz.<\/span><\/p>\n E\u015fle\u015ftirilmi\u015f \u00f6rnekler t-testi,<\/a> bir \u00f6rnekteki her g\u00f6zlemin di\u011fer \u00f6rnekteki bir g\u00f6zlemle ili\u015fkilendirilebildi\u011fi iki pop\u00fclasyon ortalamas\u0131n\u0131n e\u015fit olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in kullan\u0131l\u0131r.<\/span><\/p>\n \u00d6rne\u011fin, bir profes\u00f6r\u00fcn iki farkl\u0131 \u00e7al\u0131\u015fma y\u00f6nteminin s\u0131navlarda farkl\u0131 ortalama notlara yol a\u00e7\u0131p a\u00e7mad\u0131\u011f\u0131n\u0131 bilmek istedi\u011fini varsayal\u0131m.<\/span><\/p>\n Bunu test etmek i\u00e7in A Y\u00f6ntemini kullanacak 10 \u00f6\u011frenciyi i\u015fe al\u0131yor ve ard\u0131ndan bir teste giriyor. Daha sonra Y\u00f6ntem B’yi kullanan ayn\u0131 10 \u00f6\u011frencinin benzer zorlukta ba\u015fka bir teste haz\u0131rlanmas\u0131na ve girmesine izin verir.<\/span><\/p>\n Her iki \u00f6rnekte de t\u00fcm \u00f6\u011frenciler yer ald\u0131\u011f\u0131ndan bu senaryoda e\u015fle\u015ftirilmi\u015f \u00f6rnekler t-testi ger\u00e7ekle\u015ftirebiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, her \u00f6\u011frencinin puanlar\u0131n\u0131n bir pandas DataFrame’e nas\u0131l girilece\u011fini ve ard\u0131ndan e\u015fle\u015ftirilmi\u015f bir \u00f6rnek t testi ger\u00e7ekle\u015ftirmek i\u00e7in SciPy<\/strong> k\u00fct\u00fcphanesinin ttest_rel()<\/a> i\u015flevinin nas\u0131l kullan\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Sonu\u00e7tan \u015funu g\u00f6rebiliriz:<\/span><\/p>\n P de\u011feri 0,05’ten k\u00fc\u00e7\u00fck oldu\u011fundan, e\u015fle\u015ftirilmi\u015f \u00f6rnekler t-testinin bo\u015f hipotezini reddediyoruz ve iki y\u00f6ntemin farkl\u0131 ortalama s\u0131nav puanlar\u0131na yol a\u00e7t\u0131\u011f\u0131n\u0131 belirtmek i\u00e7in yeterli kan\u0131t oldu\u011fu sonucuna var\u0131yoruz.<\/span><\/p>\n A\u015fa\u011f\u0131daki e\u011fitimlerde Python’da di\u011fer genel g\u00f6revlerin nas\u0131l ger\u00e7ekle\u015ftirilece\u011fi a\u00e7\u0131klanmaktad\u0131r:<\/span><\/p>\n Python’da Ki-Kare Ba\u011f\u0131ms\u0131zl\u0131k Testi Nas\u0131l Yap\u0131l\u0131r?<\/a> A\u015fa\u011f\u0131daki \u00f6rneklerde pandas DataFrame kullan\u0131larak \u00fc\u00e7 farkl\u0131 t testinin nas\u0131l ger\u00e7ekle\u015ftirilece\u011fi g\u00f6sterilmektedir: Ba\u011f\u0131ms\u0131z iki \u00f6rnekli t testi Welch’in iki \u00f6rnekli t testi E\u015fle\u015ftirilmi\u015f \u00f6rnekler t testi \u00d6rnek 1: Pandalarda ba\u011f\u0131ms\u0131z iki \u00f6rnekli t testi \u0130ki pop\u00fclasyonun ortalamalar\u0131n\u0131n e\u015fit olup olmad\u0131\u011f\u0131n\u0131 belirlemek i\u00e7in ba\u011f\u0131ms\u0131z iki \u00f6rnekli t testi kullan\u0131l\u0131r. \u00d6rne\u011fin, bir profes\u00f6r\u00fcn iki farkl\u0131 \u00e7al\u0131\u015fma y\u00f6nteminin […]<\/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-3129","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
\u00d6rnek 1: Pandalarda ba\u011f\u0131ms\u0131z iki \u00f6rnekli t testi<\/strong><\/span><\/h3>\n
import<\/span> pandas as<\/span> pd\nfrom<\/span> scipy. stats<\/span> import<\/span> ttest_ind\n\n#create pandas DataFrame\n<\/span>df = pd. DataFrame<\/span> ({'method': ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A',\n 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B'],\n 'score': [71, 72, 72, 75, 78, 81, 82, 83, 89, 91, 80, 81, 81,\n 84, 88, 88, 89, 90, 90, 91]})\n\n#view first five rows of DataFrame\n<\/span>df. head<\/span> ()\n\n method score\n0 to 71\n1 To 72\n2 To 72\n3 to 75\n4 to 78\n\n#define samples\n<\/span>group1 = df[df[' method<\/span> ']==' A<\/span> ']\ngroup2 = df[df[' method<\/span> ']==' B<\/span> ']\n\n#perform independent two sample t-test\n<\/span>ttest_ind(group1[' score<\/span> '], group2[' score<\/span> '])\n\nTtest_indResult(statistic=-2.6034304605397938, pvalue=0.017969284594810425)\n<\/strong><\/span><\/pre>\n\n
\u00d6rnek 2: Pandalarda Welch’in t testi<\/strong><\/span><\/h3>\n
import<\/span> pandas as<\/span> pd\nfrom<\/span> scipy. stats<\/span> import<\/span> ttest_ind\n\n#create pandas DataFrame\n<\/span>df = pd. DataFrame<\/span> ({'method': ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A',\n 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B'],\n 'score': [71, 72, 72, 75, 78, 81, 82, 83, 89, 91, 80, 81, 81,\n 84, 88, 88, 89, 90, 90, 91]})\n\n#define samples\n<\/span>group1 = df[df[' method<\/span> ']==' A<\/span> ']\ngroup2 = df[df[' method<\/span> ']==' B<\/span> ']\n\n#perform Welch's t-test\n<\/span>ttest_ind(group1[' score<\/span> '], group2[' score<\/span> '], equal_var= False<\/span> )\n\nTtest_indResult(statistic=-2.603430460539794, pvalue=0.02014688617423973)\n<\/strong><\/span><\/pre>\n\n
\u00d6rnek 3: Pandas’ta e\u015fle\u015ftirilmi\u015f \u00f6rnekler t-testi<\/strong><\/span><\/h3>\n
import<\/span> pandas as<\/span> pd\nfrom<\/span> scipy. stats<\/span> import<\/span> ttest_rel\n\n#create pandas DataFrame\n<\/span>df = pd. DataFrame<\/span> ({'method': ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A',\n 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B'],\n 'score': [71, 72, 72, 75, 78, 81, 82, 83, 89, 91, 80, 81, 81,\n 84, 88, 88, 89, 90, 90, 91]})\n\n#view first five rows of DataFrame\n<\/span>df. head<\/span> ()\n\n method score\n0 to 71\n1 To 72\n2 To 72\n3 to 75\n4 to 78\n\n#define samples\n<\/span>group1 = df[df[' method<\/span> ']==' A<\/span> ']\ngroup2 = df[df[' method<\/span> ']==' B<\/span> ']\n\n#perform independent two sample t-test\n<\/span>ttest_rel(group1[' score<\/span> '], group2[' score<\/span> '])\n\nTtest_relResult(statistic=-6.162045351967805, pvalue=0.0001662872100210469)\n<\/strong><\/span><\/pre>\n\n
Ek kaynaklar<\/strong><\/span><\/h3>\n
Python’da Tek Y\u00f6nl\u00fc ANOVA Nas\u0131l Ger\u00e7ekle\u015ftirilir<\/a>
Python’da Fisher’in Kesin Testi Nas\u0131l Ger\u00e7ekle\u015ftirilir<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"