Nonparametric test for two independent groups

• Autores: Huamanchumo de la Cuba, Luis Emilio, Camero Jiménez, José William, Angulo Sánchez, Kesber
• Formato: artículo
• Fecha de Publicación: 2007
• Institución: Universidad Nacional de Ingeniería
• Repositorio: Revistas - Universidad Nacional de Ingeniería
• Lenguaje: español
• OAI Identifier: oai:oai:revistas.uni.edu.pe:article/384
• Enlace del recurso: https://revistas.uni.edu.pe/index.php/tecnia/article/view/384
• Nivel de acceso: acceso abierto
• Materia: prueba de Wilcoxon; distribución normal; observaciones empatadas; corrección por continuidad; Wilcoxon's test; normal distribution; correction for continuity; tied observations

This paper presents a new nonparametric test for two independent samples. It has been decided to make the inference, using all the possible differences between the comments of the two samples, and has used the Wilcoxon method for treatment, but with a different attitude when faced with a couple of comments that did not numerically presents difference, it not removed from the analysis. This test proves to be suitable from small samples (n = 5) and is very sensitive to detect significant differences between the populations from which the samples were taken. The most important thing in this test, it is also able to check when each sample has all his remarks equals (S12= S22 = 0), a problem which does not exceed t-Student test or Mann-Whitney test due to the correction for draws.