Inthisproblem,youwillexaminewhetherfamilyincomeafectsanindividual’slikelihoodtoenrollincollegebyanalyzingasurveyofapproximately4739highschoolseniorsthatwasconductedin1980withafollow-upsurveytakenin1986.ThisdatasetisbasedonadatasetfromRouse,CeciliaElena.“Democratizationordiversion?Theefectofcommunitycollegesoneducationalattainment.”JournalofBusiness&EconomicStatistics13,no.2(1995):217-224.Thedatasetiscollege.csvanditcontainsthefollowingvariables:•collegeIndicatorforwhetheranindividualattendedcollege.(Outcome)•incomeIsthefamilyincomeaboveUSD25,000peryear(Treatment)•distancedistancefrom4-yearcollege(in10sofmiles).•scoreTheseareachievementtestsgiventohighschoolseniorsinthesamplein1980.•fcollegeIsthefatheracollegegraduate?•tuitionAveragestate4-yearcollegetuition(in1000USD).•wageStatehourlywageinmanufacturingin1980.•urbanDoesthefamilyliveinanurbanarea?

DrawaDAGofthevariablesincludedinthedataset,andexplainwhyyouthinkarrowsbetweenvariablesarepresentorabsent.YoucanuseanytoolyouwanttocreateanimageofyourDAG,butmakesureyouembeditonyourcompiled.pdffile.Assumingthattherearenounobservedconfounders,whatvariablesshouldyouconditiononinordertoestimatetheefectofthetreatmentontheoutcome,accordingtotheDAGyoudrew?Explainyourdecisionindetail.Inyourexplanation,provideadefinitionofconfounding.

Chooseoneofthemethodologieswelearnedinclasstocalculateacausalefectunderconditionalignorability.Whatestimandareyoutargetingandwhy?Explainwhyyoumadeyourchoice,anddiscusstheassumptionsthatareneededtoapplyyourmethodofchoicetothisdataset.Stateifandwhyyouthinktheseassumptionsholdinthisdataset.Inaddition,chooseamethodtocomputevarianceestimates(i.e.,robuststandarderrorsorbootstrapping),anddiscussthereasonsbehindyourchoiceinthecontextofthisdataset.

UsingthemethodologyyouchoseinQuestionBtocontrolfortheconfoundersyouhaveselectedinQuestionA,aswellastherelevantRpackages,provideyourestimateofthecausalefectofthetreatmentontheoutcome.Usingyourvarianceestimatorofchoice,reportstandarderrorsand95%confidenceintervalsaroundyourestimates.Interpretyourresultsanddiscussboththeirstatisticalsignificanceandtheirsubstantiveimplications.Beasspecificanddetailedaspossible.