Multicollinearity refers to high correlation between independent variables in a regression model, when they are supposed to be independent among themselves. This causes problems in fitting the model, redundancy and and interpreting the results.
Multicollinearity can be dealt with by:
1. Assessing logically whether each independent variable included in a model is really unique and measuring something different from the others.
2. Determining the correlation between independent variables
3. Looking at scatter plots.
4. Eliminating the redundant independent variable(s).