KRUSKAL-WALLIS TESTİ VE FRIEDMAN TESTİNİN ALTERNATİF PARAMETRİK TEKNİKLERLE KARŞILAŞTIRILMASI VE BAZI PARAMETRİK VE PARAMETRİK OLMAYAN ÇOKLU KARŞILAŞTIRMA YÖNTEMLERİ İLE İNCELENMESİ

Yıl 2001, , 89 - 96, 15.01.2001

Serpil Ergün Bülbül

https://doi.org/10.14783/maruoneri.735494

Cited By: 3

Öz

The analysis of variance procedure w as concerned witlı whether three or more population means are equal. The data were interval-or ratio- level, and it was assumed the populations were normally distributed and the Standard deviations of those populations were equal. iVhen eilher of these assumplions is not met, we need an alternative test. The Kruskal-tVallis (K-IV) test is the noııparametric alternative to the one w ay analysis of variance F test, yvhich requires data measured on at least an interval scale. The K- W test requires data measured on only an ordinat scale and assurnes only continuous distributions. With three or more related or matched sampies, the Friedman test is appropriate; its parametric counterpart is the two-wav ANOVA without interaction.

The purpose of this paper is to compare these techniques. The sample consisted of 100 students who are attending the second ciass of the University of Marmara, Banking and Insurance Higlı School. Two parametric methods (one way ASOVA and two-way ANOVA without interaction) and two non-parametric methods (K- W test and Friedman test) applied on the data, The results of the nonparametric tests compared with those from the parametric tests. Also, the sample data given in the study was analyzed using multiple comparisons tests. I

Anahtar Kelimeler

parametrik analiz, kruskal-wallis testi

Kaynakça

• [1] Levin. R.J. Statistics in Research and Development. Second edt.. Chapman & Hail, London. 1991, s.607.
• [2] Mason. R.D.; Douglas A. Lind. Statistical Techniques in Business and Economics, seventh edt., Richard D. lrwin, İne.. 1990. s.656.
• [3] Conover, W.J. Practical Nonparametric Statistics, John Wiley & Sons, İne.. Nevv York, 1971, s.257.
• [4j Siegel, Sidnev. Davranış Bilimleri İçin Parametrik Olmayan İstatistikler, Çev. Yurdal Topsever. Ankara Üniversitesi Basımevi, A.Ü. Dil ve Tarih Coğrafya Fakültesi Yayınları No: 274, Ankara, 1977, ss. 184 - 185.
• [5] Daniel, W.W., J.C.Terrell, Business Statistics, Houghton Mifflin Companv, 4th edition, Boston, 1986, ss.572-573.
• [6] Orhunbilge. Nevran. Örnekleme Yöntemleri ve Hipotez Testleri, İşletme Fakültesi Yayın No:270, İstanbul, 1997, s.188.
• [7] Gibbons.Jean Dickinson. Nonparametric Statistics, Sage Publications, California, 1993, s. 1.
• [8] Özdemir, Kazım. Paket Programlar İle İstatistiksel Veri Analizi 1, 2. Baskı, Kaan Kitabevi, Eskişehir, 1999, ss.377-378.
• [9] Bülbül, Serpil Ergün. "Çoklu Karşılaştırma Testleri ve Bir Örnek Uygulama” Öneri Dergisi, Yıl: 6, Sayı: 14, Cilt: 3, Haziran 2000, İstanbul, s.97.