TY - JOUR
T1 - ¿Afecta la forma de puntuación la estructura factorial del GHQ-12? Estudio exploratorio en estudiantes iberoamericanos
AU - Urzúa, Alfonso
AU - Caqueo-Urízar, Alejandra
AU - Bargsted, Mariana
AU - Irarrázaval, Matías
N1 - Publisher Copyright:
© 2015 Fundacao Oswaldo Cruz. All rights reserved.
PY - 2015/1/1
Y1 - 2015/1/1
N2 - This study aimed to evaluate whether the scoring system of the General Health Questionnaire (GHQ-12) alters the instrument’s factor structure. The method considered 1,972 university students from nine Ibero American countries. Modeling was performed with structural equations for 1, 2, and 3 latent factors. The mechanism for scoring the questions was analyzed within each type of structure. The results indicate that models with 2 and 3 factors show better goodness-of-fit. In relation to scoring mechanisms, procedure 0-1-1-1 for models with 2 and 3 factors showed the best fit. In conclusion, there appears to be a relationship between the response format and the number of factors identified in the instrument’s structure. The model with the best fit was 3-factor 0-1-1-1-formatted, but 0-1-2-3 has acceptable and more stable indicators and provides a better format for two- and three-dimensional models.
AB - This study aimed to evaluate whether the scoring system of the General Health Questionnaire (GHQ-12) alters the instrument’s factor structure. The method considered 1,972 university students from nine Ibero American countries. Modeling was performed with structural equations for 1, 2, and 3 latent factors. The mechanism for scoring the questions was analyzed within each type of structure. The results indicate that models with 2 and 3 factors show better goodness-of-fit. In relation to scoring mechanisms, procedure 0-1-1-1 for models with 2 and 3 factors showed the best fit. In conclusion, there appears to be a relationship between the response format and the number of factors identified in the instrument’s structure. The model with the best fit was 3-factor 0-1-1-1-formatted, but 0-1-2-3 has acceptable and more stable indicators and provides a better format for two- and three-dimensional models.
KW - Questionnaires
KW - Statistical factors analysis
KW - Statistical models
UR - http://www.scopus.com/inward/record.url?scp=84937396676&partnerID=8YFLogxK
U2 - 10.1590/0102-311X00122913
DO - 10.1590/0102-311X00122913
M3 - Article
C2 - 26200377
AN - SCOPUS:84937396676
SN - 0102-311X
VL - 31
SP - 1305
EP - 1312
JO - Cadernos de Saude Publica
JF - Cadernos de Saude Publica
IS - 6
ER -