I am learning multiple regression at the momemnt. And maybe I'm just sleep deprived, but I am supposed to assess whether increasing student teacher ratios (total enrollment/teachers, str) will improve test scores. As new variables added, the T-score decreases and standard error increases. Eventually with the final variable, expenditures per student makes the standard error larger than the coefficient, resulting in the confidence interval crossing over zero and p>0.05. Teacher student ratios and expenditures per student are obviously correlated as hiring new teachers would increase expenditures per student and a correlation analysis shows r2 = -0.620 (the student teacher ratio is set up inverted so more spending on teachers would mean less students per teacher). Is the answer for factors affecting overall statistical conclusion validity, internal validity, and external validity.
.