Abnormally wide confidence intervals in logistic regression:: interpretation of statistical program results

Jokin de Irala ^[1] ; Rafael Fernandez-Crehuet Navajas ^[2] ; Amparo Serrano del Castillo ^[1]
1. [1] Universidad de Córdoba
  
  Universidad de Córdoba
  
  Cordoba, España
2. [2] Hospital Reina Sofía
  
  Hospital Reina Sofía
  
  Tudela, España
Localización: Revista Panamericana de Salud Pública = Pan American Journal of Public Health, ISSN-e 1680-5348, ISSN 1020-4989, Vol. 2, Nº. 4 (Octubre), 1997, págs. 268-271
Idioma: varios idiomas
Títulos paralelos:
- Intervalos de confianza anormalmente amplios en regresión logística:: interpretación de resultados de programas estadísticos271
Enlaces
- Texto completo
Resumen
- This study describes the behavior of eight statistical programs (BMDP, EGRET, JMP, SAS, SPSS, STATA, STATISTIX and SYSTAT) when performing a logistic regression with a simulated data set that contains a numerical problem created by the presence of a cell value equal to zero. The programs respond in different ways to this problem. Most of them give a warning, although many simultaneously present incorrect results, among which are confidence intervals that tend toward infinity. Such results can mislead the user. Various guidelines are offered for detecting these problems in actual analyses, and users are reminded of the importance of critical interpretation of the results of statistical programs