A Multinomial Logistic Regression Model for Analyzing Attitudes towards Political Activities: A Case Study in Erbil/ Kurdistan Region of Iraq

Khwazbeen Saida Fatah

College of Science, Salahaddin University

DOI: https://doi.org/10.17656/jzs.10248

Abstract: 

      Regression models are considered as the most commonly used statistical analysis techniques to describe the functional relationships between a dependent variable (either continuous or categorical) and a set of independent variables based on samples from a particular population. In this paper, a Multinomial Logistic Regression Model is proposed to investigate the variations of multi ethnic-religious people towards political attitudes. This model is applied specifically to a case study conducted in Erbil in Kurdistan Region of Iraq, where multi-ethnic and religious groups live in. Results for statistical analysis show that the variations of ethnicity or religion have no much effect on the political attitudes for the majority of citizens of this region. 

Keywords: Regression Analysis, Logistic Regression, Multinomial Logistic Regression,  

References

1-Al-Habib, A. (2012) “An Application on Multinomial Logistic Regression Model”, Pakistan Journal of Statistics & Operation Research, 8 (2), 271-298 

2-Allenby, M. and Leuk, J. (1994) ‘Modeling household purchase behaviours with logistic normal regression’, Journal of the American  Statistical Association, 89( 428), 1218– 1231; 

3-Balabanis, G. and Vassileoiu, S. (1999) “Some additional predictors of home shopping through the Internet”, Journal of Marketing  Management, 15(1), 361–385; 

4-Berry, D. (1993) Understanding Regression Assumptions, Sag university paper series on quantitative applications in the social sciences,  Beverly Hills, CA:Sage; 

5-Berry, D. And Feldman, S. (1985) Multiple Regression in Practice, Sag university paper series on quantitative applications in the social    sciences, Beverly Hills, CA:Sage; 

6-Colin, B. And Robert, J. (1984) “Calculation of polychotomous logistic regression parameters using individualized regressions”, Biometrica,  7(1), 11-18; 

7-Cornfield, J., Gordon T. and Smith W. (1961) “ Quantal response curves for experimentally uncontrolled variables, Bulletin of the  International Statistical Institute, 38 (1), 97—115; 

8-Field, A. (2009) Discovering Statistics Using SPSS, (2nd Edition), London, SAGE Publications; 

9-Henn, M., Weinstein, M. and Wring, D. (2002) “A generation apart? Youth and political participation in Britan”, British Journal of Politics and  International Relations, 4(1), 167-192; 

10-Hosmer, W. and Lemeshow, S (2000) Applied logistic regression, 2nd edition, New York, Wiley- USA: John Wiley and Sons; 

11-Li, Y. and Marsh, D. (2008) “New forms of political participation: searching for expert citizens and every day makers”, British Journal of  Politics and International Relations, 38(1), 247-272;  

12-Tabachnick, G., Fidell, S. and Osterlind, J (2001) Using multivariate statistics, US, Allyn and Bacon Boston; 

13-Thompson, B. (2008) Foundations of Behavioral Statistics, USA, New Age International Limited; 

14-Wackerly, D. Mendenhall, W. And Scheaffer, R. (2002) Mathematical Statistics with Applications, Sixth Edition, USA, Duxbury-Thomson  Learning; 

15-Walter S. And Duncan D (1967) “ Estimation of the probability of an event as a function of several variables”, Biometrika, 54(1), 79-167; 

16-Wani, J. (1971) Probability and Statistical Inference, USA, Meredith Corporation. 

View Full Article