H1: Advertisement (IV) has positive impact on Sale (DV)
Above is hypothesis of my study. Results (values) of my data are:
Coefficient: -0.56, t value -0.129, p value 0.864
P value is showing insignificant relationship but coefficient and t values are showing the reverse (negative) impact of IV on DV.
My query is that should I ignore reverse sign or I have to provide justification of negative sign?
Hello, Basically, if your variables are positively correlated, then the coefficients will be negatively correlated, which can lead to a wrong sign on one of the coefficients.
One check would be to perform a principal components regression or ridge regression. This reduces the dimensionality of the regression space, handling the multicollinearity. You end up with biased estimates but a possibly corrected signs. Whether you go with those particular results or not, it's a good diagnostic check. If you still get sign changes, it may be theoretically interesting.
A high value of p-value is indicate the likelihood of a high error in the calculation of the correlation coefficient. May be more precisely, at the current moment, based on your data, you do not have sufficient grounds to assert that there is a strong negative correlation between the variables, i.e. a lot of noise in the source data. The solution to this problem may be an increase in the number of measurements. This will decrease of the p-value and accordingly make the estimate of the coefficient more accurate. It is possible that the value of the coefficient will be change also. Possibly you should use other methods of analysis. Sometimes a simple transformation of the source data improves the correlation between variables.
The simplest answer is: Advertisement (IV) has no impact on Sale (DV). There is no need to interpret the negative coefficient because p > .05.
If P value shows insignificant relationship , it is unremarkable and no need to interpret it.
Hello, Basically, if your variables are positively correlated, then the coefficients will be negatively correlated, which can lead to a wrong sign on one of the coefficients.
One check would be to perform a principal components regression or ridge regression. This reduces the dimensionality of the regression space, handling the multicollinearity. You end up with biased estimates but a possibly corrected signs. Whether you go with those particular results or not, it's a good diagnostic check. If you still get sign changes, it may be theoretically interesting.
Karen A. Darbinyan thank you, I will again check correlation and multicollinearity .
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation