# A little clarification on linear regression

#### JSG1111

##### New Member
Hi all! I'm new to this forum so bare with me,
I recently had a test and got a sub-par score because of a pair of questions but with everything we have gone over I feel confident that I did give the right answer. I was hoping people on here could help me out!
The question reads:"Jay Bennett calculated the regression line for average 1991 SAT scores (total math plus verbal) versus number of dollars spent per student in 1991 for New Jersey School districts and obtained a slope of .0227 and a y-intercept of 707. What average SAT result (math plus verbal) does this regression line predict for students in a district that spends $2,000 per student?" a) 934 b) 752 c) 480 d) 467 e) More information is needed to make this calculation I chose e because without knowing which variable is response and which variable is explanatory you cant calculate a regression line with two unconnected variables and predict values from one another. The question also says versus which doesn't tell us anything. I understand how he got to answer b but we are assuming that this regression line predicts SAT based on district spending, what is there to say that this line was actually used for predicting district money spent based on SAT scores? Am I wrong? Any help is much appreciated. #### hlsmith ##### Not a robit I get where you are coming from since it never stated SAT regressed on Dollars, but it is indirectly implied in the question and the following line helps to support that, "What average SAT result (math plus verbal) does this regression line predict for students in a district that spends$2,000 per student?"

given the "predict" term in the sentence. If the teacher is a pushover perhaps you may make some progress, but I would imagine it comes down to if others were confused by the language as well.

#### Buckeye

##### Member
To add to the hlsmith's comment, If you see any wording such as SAT score versus dollar spent it indicates Y versus X.