The relationship between explanatory or independent variable also having an impact on the relationship of dependent and independent variable. To address such relationship, we use interaction term in our model to explain in depth the relationship and also strengthen the explanatory power of our model.
Eg. Crop (DV) and Irrigation (IV), Rainfall; Moderating variable.
The Co-relationship of rainfall and Irrigation also has an impact on the relationship of crop and Irrigation.
Interaction of independent variables in a regression model means that the effect of one variable depends on the value of one or more of the other variables. It is a common situation. Interaction is usually represented by a term consisting of a product of two or more independent variables. If the coefficient of the interaction term is significantly different from zero, it means that the effect of one variable depends on the value of the other. The model remains linear with respect to all regression coefficients, so all usual regression computational techniques should be used including checking for multicollinearity of independent variables.
The model with the interaction term X*Zis usually y = a0 + a1*bX + a2*Z + a3*X*Z + e (main effects + interaction) and not just Y= a0 + a3*X*Z + e (the interaction), and you can then use summary statistics to see if the interaction is an improvement over a simpler model with main effects only
It is possible to used with other statistical tools to see the effect of differant other variables where values are more than two.
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Agronomy
- Crop Science
- Crop