the answer is pretty exciting and searc hed some links for you to provide an answer for your question. Have fun reading them and I hope it will help you a lot:
the answer is pretty exciting and searc hed some links for you to provide an answer for your question. Have fun reading them and I hope it will help you a lot:
ANOVA is a technique for partitioning the total variation in an observed response variable into components that are due to various "controllable" or "modifiable" factors, and a residual "error" component. Here the focus is on explaining the variation due to various factors.
Linear regression is more general and versatile technique. It can be used for multiple purposes: to understand the relationship between a single response variable and multiple explanatory variables which can be categorical or continuous (regression); to develop a prediction model for the response variable (prediction); to evaluate the impact of an explanatory variable while adjusting for confounding variables (causal inference - under strict assumptions), and so on.
Regression approach is readily extended to different types of responses: continuous (linear regression), binary (logistic regression), categorical (multinomial), counts (Poisson), censored time data (Cox PH regression), etc. ANOVA, on the other hand, is limited to continuous (approximately normally distributed) variables.