If a regression has low R2 and high T stats, can we still use coefficients and intercepts to make conclusion about the dependent variable?
for Reading 11 EOC 26 we are making a conclusion with an R squared of 0.0211, SEE of 0.0710 but with coefficient and intercept with significant t-stat?
Yes, low R squared is not the be all and end all. Highly significant T-Stats are still relevant.
Are you asking if you can still interpret the coefficients in the case of a low R-square and significant t-statistics? Yes, you could, although it is bad practice to look at all of the t-statistics, given that you could look at the global F-test to see if the model has any statistically significant predictors. If you have specific theory, or you want to conduct a certain test, then a t-test is appropriate.
Assuming none of the assumptions are violated, R-square can be low in the presence of statistically significant independent variables. If you interpret some coefficients, you might find out that there is a statistical relationship, but the practical importance is minimal.