Coefficient Alpha
Background: During the 20th century the alpha coefficient (α) was widely used in the estimation of the internal consistency reliability of test scores. After misuses were identified in the early 21st century alternatives became widespread, especially the omega coefficient (ω). Nowadays, α is re-emer...
| Autores: | , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2023 |
| País: | España |
| Institución: | Universitat Autònoma de Barcelona |
| Repositorio: | Dipòsit Digital de Documents de la UAB |
| Idioma: | español |
| OAI Identifier: | oai:ddd.uab.cat:273261 |
| Acceso en línea: | https://ddd.uab.cat/record/273261 https://dx.doi.org/urn:doi:10.7334/psicothema2022.321 |
| Access Level: | acceso abierto |
| Palabra clave: | Reliability-SEM Internal consistency Reliability Cronbach's alpha Coefficient alpha Coefficient omega Congeneric measures Tau-equivalent measures Confirmatory factor analysis Software Consistencia interna Fiabilidad Alfa de Cronbach Coeficiente alfa Coeficiente omega Medidas congenéricas Medidas tau-equivalentes Análisis factorial confirmatorio |
| Sumario: | Background: During the 20th century the alpha coefficient (α) was widely used in the estimation of the internal consistency reliability of test scores. After misuses were identified in the early 21st century alternatives became widespread, especially the omega coefficient (ω). Nowadays, α is re-emerging as an acceptable option for reliability estimation. Method: A review of the recent academic contributions, journal publication habits and recommendations from normative texts was carried out to identify good practices in estimation of internal consistency reliability. Results: To guide the analysis, we propose a three-phase decision diagram, which includes item description, fit of the measurement model for the test, and choice of the reliability coefficient for test score(s). We also provide recommendations on the use of R, Jamovi, JASP, Mplus, SPSS and Stata software to perform the analysis. Conclusions: Both α and ω are suitable for items with approximately normal distributions and approximately unidimensional and congeneric measures without extreme factor loadings. When items show non-normal distributions, strong specific components, or correlated errors, variants of ω are more appropriate. Some require specific data gathering designs. On a practical level we recommend a critical approach when using the software. |
|---|