Research Article
BibTex RIS Cite

Covariates of Unit Nonresponse Error Based on Proxy Response from Household Surveys

Year 2012, Volume: 9 Issue: 1, 53 - 64, 13.07.2012

Abstract

Unit nonresponse error and its related covariates are examined from the results of a sample survey. A procedure is proposed to study unit nonresponse when data are from a two stage household sample survey in which household are the units of the first level and individuals are the units of second level. The individual person responses within the sample survey did not contain information on the nonrespondents. Therefore, household schedule variables which are based on proxy person response information are combined with the binary dependent response/nonresponse variable from the individual survey records. The idea is to estimate a logistic model whose dependent variable is the binary unit response indicator and where individual characteristics at the right hand side are approximated by household information collected at the first level. Among other models, a binary logistic regression model is proposed and the results are analyzed and interpreted by the computed odds ratios. The results have indicated several significant covariates for the model of nonresponse.

References

  • Agresti, A., (2002). Categorical Data Analysis. 2nd edition. John Wiley, Hoboken, NJ.
  • Allen, J. and Le, H., (2008). An Additional Measure of Overall Effect Size for Logistic Regression Models. Journal of Educational and Behavioral Statistics 33, 416 - 441.
  • Ayhan, H. Ö., (1981). Sources of Nonresponse Bias in 1978 Turkish Fertility Survey. Turkish Journal of Population Studies 2-3,104-148.
  • Ayhan, H. Ö., (1998). Survey Nonresponse Models and Applications in Turkey. In Official Statistics in a Changing World. Stockholm: Statistics Sweden Press, pp. 155-158.
  • Chatterjee, N., (2004). A Two Stage Regression Model for Epidemiological Studies with Multivariate Disease Classification Data. Journal of the American Statistical Association 99, 127 - 138.
  • Cox, D. R. and Oakes, D., (1984). Analysis of Survival Data. Chapman and Hall, London.
  • Hosmer, D. W., and Lemeshow, S., (1980). A Goodness-of-Fit Test for the Multiple Logistic Regression Model. Communications in Statistics A 10, 1043-1069.
  • Hosmer, D. W., and Lemeshow, S., (2000). Applied Logistic Regression (Second Edition). New York: John Wiley and Sons, Inc.
  • HUIPS., (2004). Turkey Demographic and Health Survey, 2003. Hacettepe University Institute of Population Studies, Ministry of Health General Directorate of Mother and Child Health and Family Planning, State Planning Organization and European Union. Ankara, Turkey. 309 pp.
  • Le, H. and Marcus, J., (2012). The Overall Odds Ratio as an Intuitive Effect Size Index for Multiple Logistic Regression: Examination of Further Refinements. Educational and Psychological Measurement 76(6),1001 - 1014.
  • Lemeshow, S., and Hosmer, D. W., (1982). The Use of Goodness-of- Fit Statistic in the Development of Logistic Regression Models. American Journal of Epidemiology 115, 92-106.
  • Lemeshow, S., and Hosmer, D. W., (1983). Estimation of Odds Ratio with Categorical Scaled Covariates in Multiple Logistic Regression Analysis. American Journal of Epidemiology 119,147-151.
  • Lyles, R. H., Guo, Y., and Greenland, S., (2012). Reducing Bias and Mean Squared Error Associated with Regression-Based Odds Ratio Estimators. Journal of Statistical Planning and Inference 142, 3235-3241.

Hanehalkı Araştırmalarında Yerine Cevaplayıcıdan Elde Edilen Birim Cevaplanmama Hatası Ortak Değişkenlerinin Bileşenleri

Year 2012, Volume: 9 Issue: 1, 53 - 64, 13.07.2012

Abstract

Birim cevaplanmama hatası ve ortak değişkenlerinin bileşenleri, yapılan bir örneklem araştırmasının sonuçlarına dayanarak incelenmiştir. Birinci aşaması hanehalkı ve ikinci aşaması kişi düzeyinde gerçekleşen iki aşamalı bir çalışmanın verilerde birim cevaplanmama hatasını çalışmak için bir prosedür önerilmiştir. Bu çalışmadaki kişi düzeyinde cevaplanmama ile ilgili bilgiler bulunmamaktadır. Bu nedenle, hanehalkı araştırmasında bulunan seçilmiş değişkenlerle ilgili bilgiler yerine cevaplayıcıdan elde edilmiş ve bu bilgiler aynı kişiye ait olan kişi araştırmasının sonuçlarındaki ikili cevaplama/cevaplamama bağımlı değişkeniyle birleştirilmiştir. Düşünce, cevaplanmama göstergelerini açıklamak için lojistik regresyon modeli geliştirilmesi ve modelin sağ tarafı kişi özelliklerinin ilk aşamada toplanan hanehalkı bilgileriyle yakınsamalarıdır. Diğer modellerin yanında, bir lojistik regresyon önerilmiş ve sonuçlar hesaplanan ihtimaller oranı ile analiz edilmiş ve yorumlanmıştır. Elde edilen sonuçlar, cevaplanmama modelini etkileyen bazı önemli ortak değişkenlerin mevcut olduğunu göstermektedir.

References

  • Agresti, A., (2002). Categorical Data Analysis. 2nd edition. John Wiley, Hoboken, NJ.
  • Allen, J. and Le, H., (2008). An Additional Measure of Overall Effect Size for Logistic Regression Models. Journal of Educational and Behavioral Statistics 33, 416 - 441.
  • Ayhan, H. Ö., (1981). Sources of Nonresponse Bias in 1978 Turkish Fertility Survey. Turkish Journal of Population Studies 2-3,104-148.
  • Ayhan, H. Ö., (1998). Survey Nonresponse Models and Applications in Turkey. In Official Statistics in a Changing World. Stockholm: Statistics Sweden Press, pp. 155-158.
  • Chatterjee, N., (2004). A Two Stage Regression Model for Epidemiological Studies with Multivariate Disease Classification Data. Journal of the American Statistical Association 99, 127 - 138.
  • Cox, D. R. and Oakes, D., (1984). Analysis of Survival Data. Chapman and Hall, London.
  • Hosmer, D. W., and Lemeshow, S., (1980). A Goodness-of-Fit Test for the Multiple Logistic Regression Model. Communications in Statistics A 10, 1043-1069.
  • Hosmer, D. W., and Lemeshow, S., (2000). Applied Logistic Regression (Second Edition). New York: John Wiley and Sons, Inc.
  • HUIPS., (2004). Turkey Demographic and Health Survey, 2003. Hacettepe University Institute of Population Studies, Ministry of Health General Directorate of Mother and Child Health and Family Planning, State Planning Organization and European Union. Ankara, Turkey. 309 pp.
  • Le, H. and Marcus, J., (2012). The Overall Odds Ratio as an Intuitive Effect Size Index for Multiple Logistic Regression: Examination of Further Refinements. Educational and Psychological Measurement 76(6),1001 - 1014.
  • Lemeshow, S., and Hosmer, D. W., (1982). The Use of Goodness-of- Fit Statistic in the Development of Logistic Regression Models. American Journal of Epidemiology 115, 92-106.
  • Lemeshow, S., and Hosmer, D. W., (1983). Estimation of Odds Ratio with Categorical Scaled Covariates in Multiple Logistic Regression Analysis. American Journal of Epidemiology 119,147-151.
  • Lyles, R. H., Guo, Y., and Greenland, S., (2012). Reducing Bias and Mean Squared Error Associated with Regression-Based Odds Ratio Estimators. Journal of Statistical Planning and Inference 142, 3235-3241.
There are 13 citations in total.

Details

Primary Language English
Subjects Statistics
Journal Section Research Articles
Authors

Ahmet Sinan Türkyılmaz

H. Öztaş Ayhan This is me

Publication Date July 13, 2012
Published in Issue Year 2012 Volume: 9 Issue: 1

Cite

APA Türkyılmaz, A. S., & Ayhan, H. Ö. (2012). Covariates of Unit Nonresponse Error Based on Proxy Response from Household Surveys. İstatistik Araştırma Dergisi, 9(1), 53-64.