Assume that we ran an experiment on providing free access to netflix to college students and we wanted to know how it will impact their exam scores.

1. Assume that we ran an experiment on providing free access to netflix to college students and we wanted to know how it will impact their exam scores. We gave 400 student free access to Netflix and 400 students did not get it for free.

a. If the average exam score for treatment students is 85 and the average exam score for control students is 80, while the standard deviation for each group is 60, what would the 95% confidence intervals be for the average of each group? (To be accurate please use 1.96 SE's to compute the intervals).

1. Is there a statistically significant difference between the two groups? Why?
2. Imagine instead that the standard deviation for each group was 40. What would be
3. the lowest grade possible (whole number) that the treatment group could get where we would still be able to detect a statistically significant difference?