Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova

Volume: 14 Number: 1 January 1, 2014
EN TR

Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova

Abstract

The purpose of this study was to examine the effect of missing data handling methods on the parameters of t-test and ANOVA. The study was conducted with simuated data sets. These data sets were produced in a way that they would have normal distributions in high and low correlation and their sizes were 50, 100, 200, 400 units. Under random conditions, data sets were reduced %5, %10, %20 in the form of MCAR. In the simulated data sets, mean substitution method, regression method, expectation-maximization (EM) method and deletion method were applied. Results showed that in different sample sizes and correlations, findings were differentiated. It is observed that in data sets with low sample sizes, regression and EM application were usefull on the other hand in data sets with larger sample sizes, mean substitution method instead of regression method had more consistent results

Keywords

References

  1. Afiffi A. And Elashoff R. M. (1966). Missing observations in multivariate statistics: ı. review of the literature, Journal of the American Statistical Association, 61(315) 595-604.
  2. Allison P. D. (2001). Missing data, sage university papers series on quantitativ eapplications in the social sciences, ThousandsOaks, CA, Sage.
  3. Alpar, R. (2003). Uygulamalı çok değişkenli istatistiksel yöntemlere giriş-1, Nobel Kitabevi.
  4. Bal C. (2003). Çok gruplu veri setlerinde kayıp gözlem sorununun çözümlenmesi ve sağlık alanında bir uygulama, Doktora Tezi,Eskişehir Osmangazi Üniversitesi Sağlık Bilimleri Enstitüsü, Eskişehir.
  5. Baygül A, (2007). Kayıp veri analizinde sıklıkla kullanılan etkin yöntemlerin değerlendirilmesi, Yüksek Lisans Tezi, İstanbul Üniversitesi Sağlık Bilimleri Enstitüsü, İstanbul.
  6. Cheema, J. (2012). Handling missing data in educational research using spss. Unpublished doctoral dissertation. George Mason University, USA.
  7. Çokluk Ö. , Kayrı M., (2011). Kayıp değerlere yaklaşık değer atama yöntemlerinin ölçme araçlarının geçerlik ve güvenirliği üzerindeki etkisi, Kuram ve Uygulamada Eğitim Bilimleri, Kış; 289-309.
  8. Dempster, A. P.,Laird, N. M, and Rubin, D. (1977). Maximum likelihood estimation from incomplete data via the EM algorithm, JournalRoyal Statistical Soc. 39

Details

Primary Language

English

Subjects

-

Journal Section

-

Publication Date

January 1, 2014

Submission Date

January 28, 2015

Acceptance Date

-

Published in Issue

Year 2014 Volume: 14 Number: 1

APA
Köse, İ. A., & Öztemur, B. (2014). Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova. Abant İzzet Baysal Üniversitesi Eğitim Fakültesi Dergisi, 14(1), 400-412. https://doi.org/10.17240/aibuefd.2014.14.1-5000091519
AMA
1.Köse İA, Öztemur B. Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova. BAİBÜEFD. 2014;14(1):400-412. doi:10.17240/aibuefd.2014.14.1-5000091519
Chicago
Köse, İbrahim Alper, and Begüm Öztemur. 2014. “Examining the Effect of Missing Data Handling Methods on the Parameters of T-Test and Anova”. Abant İzzet Baysal Üniversitesi Eğitim Fakültesi Dergisi 14 (1): 400-412. https://doi.org/10.17240/aibuefd.2014.14.1-5000091519.
EndNote
Köse İA, Öztemur B (January 1, 2014) Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova. Abant İzzet Baysal Üniversitesi Eğitim Fakültesi Dergisi 14 1 400–412.
IEEE
[1]İ. A. Köse and B. Öztemur, “Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova”, BAİBÜEFD, vol. 14, no. 1, pp. 400–412, Jan. 2014, doi: 10.17240/aibuefd.2014.14.1-5000091519.
ISNAD
Köse, İbrahim Alper - Öztemur, Begüm. “Examining the Effect of Missing Data Handling Methods on the Parameters of T-Test and Anova”. Abant İzzet Baysal Üniversitesi Eğitim Fakültesi Dergisi 14/1 (January 1, 2014): 400-412. https://doi.org/10.17240/aibuefd.2014.14.1-5000091519.
JAMA
1.Köse İA, Öztemur B. Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova. BAİBÜEFD. 2014;14:400–412.
MLA
Köse, İbrahim Alper, and Begüm Öztemur. “Examining the Effect of Missing Data Handling Methods on the Parameters of T-Test and Anova”. Abant İzzet Baysal Üniversitesi Eğitim Fakültesi Dergisi, vol. 14, no. 1, Jan. 2014, pp. 400-12, doi:10.17240/aibuefd.2014.14.1-5000091519.
Vancouver
1.İbrahim Alper Köse, Begüm Öztemur. Examining the Effect of Missing Data Handling Methods on the Parameters of t-Test and Anova. BAİBÜEFD. 2014 Jan. 1;14(1):400-12. doi:10.17240/aibuefd.2014.14.1-5000091519

Cited By