Question

Consider the following program. Suppose following the EC395 , I offer private tutoring the students in the bottom fifth of the class in the grade distribution. To see whether my tutoring had a positive effect on the final grade, I estimate the following regression:

FinalGradei = β0 + β1Tutoringi + ui

where Tutoring is an indicator variable that takes the value 1 if the student received tutoring from me and 0 otherwise. Critically discuss whether this study is likely to be internally valid and whether it is likely to be externally valid. [5 marks]

Answer 1

Internal validity refers to a concept where it showa the cause and effect relationship relationship between the outcome and the treatment. Whereas, external validity is the generalization of the results of a study that has been carried out.

In the above regression equation,

FinalGradei = β0 + β1Tutoringi + ui

Where, FinalGradei = dependent variable and β0 + β1Tutoringi + ui = independent variable

So, to see whether my tutoring had a positive effect on the final grade , we analyse the cause and effect relationship that is likely to present in internal validity. So, confidence that changes in dependent variable are actually caused by the independent variable. Therefore, this study is likely to be internally valid.