Résumé
Analyse de corrélation : fondements, démarche, intérêt et limites
Auteur (s) : José Mangalu Mobhe Agbada et Eugenie Kabali Hamuli
Abstract :
In this paper, the authors examine the relationship between two quantitative variables using the correlation test, a method that is commonly used but often misapplied and misinterpreted. They propose to revisit, in a new and contextualized light, not only the theoretical presuppositions and conditionalities relating to the application of this test, but also the interest and pitfalls to be avoided in its use and in the interpretation of the results resulting from it. Correlation analysis seeks to check whether there is a linear link between variables and whether this link is strong enough not to be due to chance. The article also presents concrete examples to allow readers to put the concepts into practice, highlighting the importance of mastering this technique, which is often a prerequisite for the use of more advanced statistical tests, such as linear regression.
Editeur : RECOSH
Télécharger l'article complet ici