Research Article


DOI :10.26650/ekoist.2021.35.972604   IUP :10.26650/ekoist.2021.35.972604    Full Text (PDF)

Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality

Sibel SelimGizem Kılınç

Infant mortality is one of the most important indicators of development and varies across countries. It is thus important to measure this difference statistically and econometrically. This study examines the factors affecting the number of infant deaths among married women between the ages of 15 and 49 through different count data regression models using 2018 Turkey Demographic and Health Survey (TDHS) Syrian Migrant and Turkey Sample data—implemented by the Hacettepe University Institute of Population Studies. Various count data regression models were compared to determine the best analysis method for this study where infant deaths were expressed as a count variable. In the presence of overdispersion, the Negative Binomial regression model was determined to be the most suitable model after comparing the Poisson, Negative Binomial, Zero-Inflated Poisson, and Zero-Inflated Negative Binomial regression models. The results of this study showed that the most important variables affecting infant mortality were the size of the household, the duration of marriage, the desire for more children, single risk factors, the use of birth control methods, the total number of stillbirths, multiple births, and migration numbers.

DOI :10.26650/ekoist.2021.35.972604   IUP :10.26650/ekoist.2021.35.972604    Full Text (PDF)

Bebek Ölümlerini Etkileyen Faktörler Üzerine Sayma Veri Regresyon Modellerinin Karşılaştırmalı Analizi

Sibel SelimGizem Kılınç

Bebek ölümlülüğü, kalkınmanın en önemli göstergelerinden biri olup ülkelerde farklılık göstermektedir. Bu farklılığı istatistiksel ve ekonometrik olarak ölçmek önemlidir. Bu çalışma, Hacettepe Üniversitesi Nüfus Etütleri Enstitüsü tarafından uygulanan 2018 yılı Nüfus ve Sağlık Araştırması (TDHS) Suriyeli göçmen ve Türkiye örneklemine ait veriler kullanılarak sayma veri regresyon modelleri ile 15-49 yaş arasındaki evli kadınların ölen bebek sayılarını etkileyen faktörleri incelemektedir. Bebek ölümlerinin sayma değişkeni olarak ifade edildiği bu çalışmada en iyi analiz yöntemini belirlemek için çeşitli sayma veri regresyon modelleri karşılaştırılmıştır. Aşırı yayılımın varlığında Poisson, Negatif Binom, Sıfır Yığılmalı Poisson ve Sıfır Yığılmalı Negatif Binom regresyon modelleri karşılaştırılarak Negatif Binom regresyon modelinin en uygun model olduğu belirlenmiştir. Bu çalışmanın sonuçları bebek ölümlerini etkileyen en önemli değişkenlerin; hanehalkı büyüklüğü, evlilik süresi, daha fazla çocuk isteği, tekli risk faktörleri, doğum kontrol yöntemi kullanımı, toplam ölü doğum sayısı, çoklu doğumlar ve göç sayısı olduğunu göstermiştir. 


EXTENDED ABSTRACT


When infant deaths in Turkey are examined, it can be seen that these deaths have decreased over time. However, this decrease is still not at the desired levels. Although Turkey is among the top developing countries in terms of manpower and economic resources, infant mortality has not shown a parallel decline. In addition to being a health indicator, infant mortality is undoubtedly one of the most important indicators of human development. There is a strong relationship between the level of development of the society and infant mortality. As the level of development increases, infant mortality decreases. It changes with regional development in countries, and it is important to measure this change statistically and econometrically. This study examines the factors affecting the number of infant deaths among married women between the ages of 15 and 49 through count data regression models using 2018 Turkey Demographic and Health Survey (TDHS) Syrian Migrant and Turkey Sample data—implemented by the Hacettepe University Institute of Population Studies— and determines the best analysis method by comparing models.

Statistical analysis of count variables has a very long history. In regression models using count data, the dependent variable takes an integer value greater than zero. Among the count data regression models, the Poisson regression model is the simplest. This model offers features where the conditional mean of the output is equal to the conditional variance. The Negative Binomial model is a special case of the Poisson regression model and is used as an alternative method in practice as a result of over-dispersion. The Zero-Inflated Poisson and Zero-Inflated Negative Binomial model is used in datasets where the value of zero is excessive and there is overdispersion. In this study where infant deaths were expressed as a count variable, the existence of over-dispersion was detected, and the Negative Binomial regression model was determined to be the most suitable model after comparing the Poisson, Negative Binomial, Zero-Inflated Poisson, and Zero-Inflated Negative Binomial regression models.

The dependent variable used in the analyses in this study was the number of infant deaths, with the ages ranging from 0 to 1. The explanatory factors affecting infant deaths were education levels, working status of the mother and father, region, place of residence, mother's age at first birth, household size, family planning, ideal number of children, relationship between the relatives, use of a contraceptive method, stillbirths, duration of marriage, risk factors for infant death, number of migrations, smoking status, sporting status, desire for more infants, and multiple births. In this study where infant deaths were expressed as a count variable, the existence of overdispersion was detected and the Negative Binomial regression model was determined to be the most suitable model after comparing the Poisson, Negative Binomial, Zero-Inflated Poisson, and Zero-Inflated Negative Binomial regression models. The findings of this study showed the differences in infant mortality for the TDHS Syrian The dependent variable used in the analyses in this study was the number of infant deaths, with the ages ranging from 0 to 1. The explanatory factors affecting infant deaths were education levels, working status of the mother and father, region, place of residence, mother's age at first birth, household size, family planning, ideal number of children, relationship between the relatives, use of a contraceptive method, stillbirths, duration of marriage, risk factors for infant death, number of migrations, smoking status, sporting status, desire for more infants, and multiple births. In this study where infant deaths were expressed as a count variable, the existence of overdispersion was detected and the Negative Binomial regression model was determined to be the most suitable model after comparing the Poisson, Negative Binomial, Zero-Inflated Poisson, and Zero-Inflated Negative Binomial regression models. The findings of this study showed the differences in infant mortality for the TDHS Syrian


PDF View

References

  • AFAD (2014). Türkiye’deki Suriyeli Kadınlar, Başbakanlık Afet ve Acil Durum Yönetimi Başkanlığı, Ankara. google scholar
  • Aile ve Sosyal Politikalar Bakanlığı Aile ve Toplum Hizmetleri Genel Müdürlüğü (2016). Suriyeliler ile Türkiye Cumhuriyeti Vatandaşları Arasındaki Evlilik İlişkileri Araştırması. İstanbul: Çizge Tanıtım. google scholar
  • Akinpelu, K. P., Yusuf, O. B., Akpa, O.M. & Gbolahan, A. O. (2016). Zero Inflated Regression Models with Application to Malaria Surveillance Data. International Journal of Statistics and Applications, 6(4), 223-234. google scholar
  • Aladağ Çiftdemir, N., Özden, D., Sayın, M., Oksay, A., Tarım, C., Vatansever Özbek, Ü., Ceylan, G., & Şahin, B. (2016). Edirne İli 2012-2013 Yılları Arası Bebek Ölümlerinin Değerlendirilmesi. The Journal of Pediatric Research, 2(4), 193-196. google scholar
  • Alpu, O., & Fidan, H. (2004). Sequential Probit Model for Infant Mortality Modelling in Turkey. Journal of Applied Sciences, 4(4), 590-595. google scholar
  • Aydın Yılmaz, E. S. (2019). Suriyeli Mülteci Kadınlar: Göç Travması ve Entegrasyon. in Eds. (Esen, A. ve Duman, M.). Türkiye’de Geçici Koruma Altındaki Suriyeliler: Tespitler ve Öneriler. İstanbul: Aryan Basım Tanıtım. google scholar
  • Barın, H. (2015). Türkiye’deki Suriyeli Kadınların Toplumsal Bağlamda Yaşadıkları Sorunlar ve Çözüm Önerileri. Göç Araştırmaları Dergisi, 1(2), 10-56. google scholar
  • Beaujean, A.A., & Morgan, G.B. (2016). Tutorial on Using Regression Models with Count Outcomes Using R. Practical Assessment Research & Evaluation, 21(2), 1531-7714. google scholar
  • Becker, G. S. (1965). A Theory of the Allocation of Time. The Economic Journal, 75(299), 493517. google scholar
  • Bird, S. T., & Bauman, K. E. (1998). State-Level Infant, Neonatal, and Postneonatal Mortality: the Contribution of Selected Structural Socioeconomic Variables. International Journal of Health Services, 28(1), 13-27. google scholar
  • Bodur, S., Durduran, Y., Küçükkendirci, H., & Doğan, C. (2009). Bebek Ölümlerinin Prenatal-Natal Sağlık Hizmet Kullanımı ve Demografik Özelliklerle İlişkisi: Vaka Kontrol Çalışması. Dicle Tıp Dergisi, 36(4), 288-293. google scholar
  • Caldwell, J. C. (1979). Education As A Factor in Mortality Decline: an Examination of Nigerian Data. Popul Stu, 33, 395-413. google scholar
  • Cameron, A.C, & Johansson, P. (1997). Count Data Regression Using Series Expansions: with Application. Journal of Applied Econometrics, 12, 203-223. google scholar
  • Cameron, A. C., & Trivedi, P. (1998). Regression Analysis of Count Data. West Nyack, NY, USA: Cambridge University Press. google scholar
  • Casterline, J. B., Cooksey, E. C., & Ismail, A. F. E. (1989). Household Income and Child Survival in Egypt. Demography, 26 (1), 15-35. google scholar
  • Civelek, Y. H. (2020). Türkiye’deki Suriyeli Göçmen Kadınlar Arasında Güvenli Doğurganlık ve Güvenli Annelik. Göç Dergisi, 7(2), 245-272. google scholar
  • Çoban, H. (2009). Sağlık Ekonomisi ve Türkiye’de Sağlık Hizmetlerinin Yeniden Yapılandırılması. Dokuz Eylül Üniversitesi, Doktora Tezi. google scholar
  • DaVanzo J., Butz W. P., & Habicht J. P. (1983). How Biological And Behavioural Influences On Mortality in Malaysia Vary During The First Year of Life. Population Studies, 37(3), 381-402. google scholar
  • Deniz, Ö. (2005). Poisson Regresyon Analizi. İstanbul Ticaret Üniversitesi Fen Bilimleri Dergisi, 4(7), 59-72. google scholar
  • Denuit, M., Marechal, X., Pitrebois, S., & Walhin, J.F. (2007). Actuarial Modelling of Claim Counts: Risk Classification, Credibilityand Bonus-Malus Systems.. England: John Wiley & Sons. google scholar
  • Diker E. (2018). Türkiye’deki Suriyeli Mültecilerle İlgili Kaynakçalar: Sağlık (MİReKOC Çalışma Notları) Koç Üniversitesi Göç Araştırmaları Uygulama ve Araştırma Merkezi. google scholar
  • Erdoğan, M. M. (2019). Türkiye’deki Suriyeli Mülteciler. Türk-Alman Üniversitesi Göç ve Uyum Araştırmaları Merkezi. google scholar
  • Ergöçmen, B. A., Hancıoğlu, A., Koç, İ., & Ünalan, T. (2004). Binyıl Kalkınma Hedeflerine Demografik Bakış. Türkiye: Hacettepe Üniversitesi Nüfus Etütleri Enstitüsü ve Birleşmiş Milletler Nüfus Fonu, Ankara. google scholar
  • Eryurt, M. A., & Koç, İ. (2009). Yoksulluk ve Çocuk Ölümlülüğü: Hanehalkı Refah Düzeyinin Çocuk Ölümlülüğü Üzerindeki Etkisi. Çocuk Sağlığı ve Hastalıkları Dergisi, 52, 113-121. google scholar
  • Filmer, D., & Pritchett, L. (1997). Child Mortality And Public Spending on Health; How Much Does Money Matter?. The World Bank Policy Research Working Paper 1864, Washington DC. google scholar
  • Folland, S., Goodman, A.C., & Stano, M. (2010). The Economics of Health And Health Care. 7th Edition, Boston: Pearson Prentice Hall. google scholar
  • Genç, D. H., & Öztürk D. (2016). Türkiye’deki Suriyeli Kadınlar: Mevcut Durumları, Sorunlar ve Çözüm Önerileri. İstanbul Medipol Üniversitesi Siyaset Bilimi ve Uluslararası İlişkiler Bölümü Liberal Düşünce Dergisi, 21(84), 67-91. google scholar
  • Greene, W. H. (1994). Accounting for Excess Zeros and Sample Selection in Poisson and Negative Binomial Regression Models. New York University, New York. google scholar
  • Grossman, M. (1972). On the Concept of Health Capital and the Demand for Health. Journal of Political Economy, 80(2), 223-255. google scholar
  • Gürsoy Tezcan, A. (1992). Infant Mortality: a Turkish Puzzle?. Health Transition Review, 2(2), 131-149. google scholar
  • Hacettepe University Institute of Population Studies, http://www.hips.hacettepe.edu.tr/tnsa2018/ rapor/ 2018_TNSA_SR.pdf s.xv. Erişim tarihi: 21.05.2020. google scholar
  • Hilbe, J. M. (2011). Negative Binomial Regression. Second Edition, Cambridge University Press. google scholar
  • Ihlamur Öner S.G. (2014). Türkiye’nin Suriyeli Mültecilere Yönelik Politikası. Ortadoğu Analiz, 6, 42-5. google scholar
  • Ismail, N., & Zamani, H. (2013). Estimation of Claim Count Data Using Negative Binomial Generalized Poisson, Zero-Inflated Negative Binomia land Zero-Inflated Generalized Poisson Regression Models. Casualty Actuarial Society E-Forum, Spring, 1-18. google scholar
  • Karaca, A. G. (2018). Sayma Verileri İçin Regresyon Modellerinin Karşılaştırılması Üzerine Bir Uygulama. Gazi Üniversitesi Fen Bilimleri Enstitüsü, Ankara. google scholar
  • Kılıç, E. (2014). Ekonomik Öncü Göstergelerden BES Katılımcı Sayısı Üzerine Etkisi. Maliye Finans Yazıları, 28(102), 65-83. google scholar
  • Lambert, D. (1992). Zero-Inflated Poisson Regression with an Application to Defects in Manufacturing. Technometrics, 34(1), 1-14. google scholar
  • Long, J. S. (1997). Regression Models for Categorical and Dependent Variables. London: Sage Publications. google scholar
  • Long, J.S. & Freese, J. (2014). Regression Models for Categorical Dependent Variables Using Stata. StataCorp LP College Station, Third Edition, Texas: A Stata Press Publication. google scholar
  • McCullah P. & Nelder, J. A. (1996). Generalized Linear Models. Chapman & Hall, Second Edition. google scholar
  • Mete, B., Yıldız, S., Söyiler, V., Doğan, E., Durmuş, H., & Timur, A. (2018). Bingöl İlinde 2016 Yılına ait Bebek Ölümleri ve Nedenlerinin İncelenmesi. Pedıatr Pact Res, 6(1), 1-5. google scholar
  • Mosley, W. H., & Chen, L. C. (1984). An Analytical Framework for the Study of Child Survival in Developing Countries. Population and Development Review,10, 25-45. google scholar
  • Myers R. H. (1990). Classical and Modern Regression with Applications. Second Edition, The Duxbury Advanced Series in Statistics and Decision Sciences. USA: PWS-KENT Publishing Company. google scholar
  • Olsen, J. A. (2009). Principles in Health Economics and Policy. Oxford, UK: Oxford University Press. google scholar
  • Peng, J. (2013). Count Data Models for Injury Data from the National Health Interview Survey. (M. Sc. Thesis), The Ohio State University Graduate Program in Public Health, Columbus. google scholar
  • Riddle, L. (1997). The Turkish Child Mortality Puzzle Continues: Evidence From the 1993 Demographic and Health Survey, Population Research Center. google scholar
  • Rosenzweig, M. R., & Schultz, T. P. (1982). Market Opportunities, Genetic Endowments, and Intrafamily Resource Distribution: Child Survival in Rural India. The American Economic Review, 72(4), 803-815. google scholar
  • Shorter, F. C., & Macura, M. (1982). Trends in Fertility and Mortality in Turkey, 1935-1975. Washington, DC: The National Academies Press. google scholar
  • Seçkin, N. (2009). Determinants of Infant Mortality in Turkey. Middle East Technical University, (Unpublished Master’s Thesis), Ankara. google scholar
  • Selim, S., & Balyaner, İ. (2017). Türkiye’de Hanehalkının Sahip Olduğu Bilişim Teknolojileri Ürünleri Sayısını Belirleyen Faktörlerin Araştırılması: Bir Sayma Veri Modeli. Mehmet Akif Ersoy Sosyal Bilimler Enstitüsü Dergisi, 9(22), 428-454. google scholar
  • Selim, S., & Üçdoğruk, Ş. (2003). Sayma Veri Modelleri ile Çocuk Sayısı Belirleyicileri: Türkiye’deki Seçilmiş İller için Sosyoekonomik Analizler. D.E.U. İ.İ.B.F. Dergisi, 18(2), 13-31. google scholar
  • Tekçe, B., & Shorter, F. C. (1984). Determinants of Child Mortality: A Study of Squatter Settlements in Jordan. Population and Development Review, 10, 257-280. google scholar
  • Özpınar, E. (2016). Bebek Ölüm Hızı Türkiye Genelinde Azalıyorken Neden Suriyeli Nüfusun Fazla Olduğu Yerlerde Artıyor?. TEPAV, Değerlendirme Notu. google scholar
  • Tezcan, S. (1985). Türkiye’de Bebek ve Çocuk Ölümleri. Hacettepe Üniversitesi Tıp Fakültesi Halk Sağlığı Anabilim Dalı, Ankara. google scholar
  • Tezcan, S. (1985). Medico-Social Causes and Preventability of Infant Deaths in Etimesgut Health District. Turkish Journal of Population Studies, 7, 43-59. google scholar
  • Tuğ, K. (2005). Socio-Demographic Analysis of Early Age Mortality in the Southeast Anatolian Region. Hacettepe University Institute of Population Studies, (Unpublished Master’s Thesis), Ankara. google scholar
  • Tüylüoğlu, Ş. & Tekin, M. (2009). Gelir Düzeyi ve Sağlık Harcamalarının Beklenen Yaşam Süresi ve Bebek Ölüm Oranı Üzerindeki Etkileri. Çukurova Üniversitesi İİBF Dergisi 13(1), 1-31. google scholar
  • Türkiye Nüfus ve Sağlık Araştırması (TNSA) (2018). Hacettepe Üniversitesi Nüfus Etütleri Enstitüsü, Türkiye Mikro Veri Seti. google scholar
  • Türkiye Nüfus ve Sağlık Araştırması (TNSA) (2018). Hacettepe Üniversitesi Nüfus Etütleri Enstitüsü, Suriyeli Göçmenler Örneklemi Mikro Veri Seti. google scholar
  • UNICEF (2019). UNICEF 2019 Yıllık Raporu. Türkiye-UNICEF Ülke İşbirliği Programı. google scholar
  • Wagstaff, A. (1986). The Demand for Health: A Simplified Grossman Model. Bulletin of Economic Research. 38, 93-95. google scholar
  • Vuong, Q. (1989). Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses. Econometrica, 57, 307-334. google scholar
  • Yanıkkaya, H., & Selim, S. (2010). The Determinants of İnfant Mortality in Turkey: A Disaggregated Analysis. İktisat İşletme ve Finans Dergisi, 25(286), 61-88. google scholar
  • Yüksel, İ. (2008). Quantitative and Qualitative Analysis of Cultural Factors Affecting Early Age Mortality in Turkey. Hacettepe University Institute of Population Studies, (Unpublished Ph. D. Thesis), Ankara google scholar
  • Yüksel, İ., & Koç, İ. (2010). Türkiye’de Bebek ve Çocuk Ölümleri Halen Bir Bilmece Mi?. Çocuk Sağlığı ve Hastalıkları Dergisi, 53, 87-97. google scholar

Citations

Copy and paste a formatted citation or use one of the options to export in your chosen format


EXPORT



APA

Selim, S., & Kılınç, G. (2021). Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality. Ekoist: Journal of Econometrics and Statistics, 0(35), 147-179. https://doi.org/10.26650/ekoist.2021.35.972604


AMA

Selim S, Kılınç G. Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality. Ekoist: Journal of Econometrics and Statistics. 2021;0(35):147-179. https://doi.org/10.26650/ekoist.2021.35.972604


ABNT

Selim, S.; Kılınç, G. Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality. Ekoist: Journal of Econometrics and Statistics, [Publisher Location], v. 0, n. 35, p. 147-179, 2021.


Chicago: Author-Date Style

Selim, Sibel, and Gizem Kılınç. 2021. “Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality.” Ekoist: Journal of Econometrics and Statistics 0, no. 35: 147-179. https://doi.org/10.26650/ekoist.2021.35.972604


Chicago: Humanities Style

Selim, Sibel, and Gizem Kılınç. Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality.” Ekoist: Journal of Econometrics and Statistics 0, no. 35 (Jun. 2022): 147-179. https://doi.org/10.26650/ekoist.2021.35.972604


Harvard: Australian Style

Selim, S & Kılınç, G 2021, 'Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality', Ekoist: Journal of Econometrics and Statistics, vol. 0, no. 35, pp. 147-179, viewed 30 Jun. 2022, https://doi.org/10.26650/ekoist.2021.35.972604


Harvard: Author-Date Style

Selim, S. and Kılınç, G. (2021) ‘Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality’, Ekoist: Journal of Econometrics and Statistics, 0(35), pp. 147-179. https://doi.org/10.26650/ekoist.2021.35.972604 (30 Jun. 2022).


MLA

Selim, Sibel, and Gizem Kılınç. Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality.” Ekoist: Journal of Econometrics and Statistics, vol. 0, no. 35, 2021, pp. 147-179. [Database Container], https://doi.org/10.26650/ekoist.2021.35.972604


Vancouver

Selim S, Kılınç G. Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality. Ekoist: Journal of Econometrics and Statistics [Internet]. 30 Jun. 2022 [cited 30 Jun. 2022];0(35):147-179. Available from: https://doi.org/10.26650/ekoist.2021.35.972604 doi: 10.26650/ekoist.2021.35.972604


ISNAD

Selim, Sibel - Kılınç, Gizem. Comparative Analysis of Count Data Regression Models on Factors Affecting Infant Mortality”. Ekoist: Journal of Econometrics and Statistics 0/35 (Jun. 2022): 147-179. https://doi.org/10.26650/ekoist.2021.35.972604



TIMELINE


Submitted07.07.2021
Accepted19.11.2021
Published Online31.12.2021

LICENCE


Attribution-NonCommercial (CC BY-NC)

This license lets others remix, tweak, and build upon your work non-commercially, and although their new works must also acknowledge you and be non-commercial, they don’t have to license their derivative works on the same terms.


SHARE




Istanbul University Press aims to contribute to the dissemination of ever growing scientific knowledge through publication of high quality scientific journals and books in accordance with the international publishing standards and ethics. Istanbul University Press follows an open access, non-commercial, scholarly publishing.