Question

5. You want to estimate the effect of hours spent in an SAT preparation course on total SAT score. You draw a random sample of high school seniors in at the end of the school year (after they have taken the SAT) from Boston area schools and ask them how many hours they spent in an SAT prep course and what SAT score they received. You set out to estimate the following regression: (a) (5 points) In equation 3, what is the expected sign of A if the course is effective? (b) (5 points) What is the interpretation of Bo? (c) (5 points) Given this is a random sample of students, some of whom voluntarily chose to take an SAT prep course, is B1 the causal effect of hours in a prep course on the SAT score? If not, is it too big or too small? Explain

Answer 1

Given that ou cont estimate the effect of houns spent in G SAT Prereaaton couse on' total SAT SCore. You set outtestimate the folluoing regrect on Ca) S equation s he expected of P the counte s effective り preferator cour ses are eFective , f, should be. acton equal, an inciease in hours Should ncge ase tSAT . P, should be +ve, ese equal, an addi honal "o the Pnepesnation courrse should impyove SLone yhe did Ave汃 SAT SCOre among s students who o houns in the reperalion lg)昆is the ttercept term 일扐e negren,on model ana Pt chouast theST SCone achieved, when the sta dent does not spend angfine on prepavafton on the SAT Pe panahi tourse at all. hu, tis the scove achiaved when the ayder des not sudy t the SAT at alhePepesatio (c f1107Hr(epresent the İmpact on the Arscore per hourr sPent on he EAT Pepeation coga . This sth slope to eficient and tho the impact on the sa scone rehou spent on the A psepesation counte The lope. Coe fficient can be terted or using the Stat t .ce e iH car. bican ( e. The ㅏ,'ghe^r the vc1b Coe#cient greate, te inpact er hout spenr on