# CH.9- Assessing Studies.pdf - Assessing Studies Based on...

• 56

Course Hero uses AI to attempt to automatically extract content from documents to surface to you and others so you can study better, e.g., in search results, to enrich docs, and more. This preview shows page 1 - 10 out of 56 pages.

Assessing Studies Based on Multiple RegressionRENNESSchool of BusinessInstructor: Saeed Mousa
A Framework for Assessing Statistical Studies:Internal and External ValidityInternal validity:the statistical inferences about causal effects are validfor the population being studiedExternal validity:the statistical inferences can be generalized from thepopulation studied to other populations
Threats to External Validity of Multiple RegressionStudiesAssessing threats to external validity requires detailed substantiveknowledge and judgment on a case-by-case basis.How far can we generalize class size results from California?Differences in populationsCalifornia in 2019?Massachusetts in 2019?Mexico in 2019?Differences in settingsdifferent legal requirements (e.g. special education)different treatment of bilingual educationDifferences in teacher characteristics
Threats to Internal Validity of Multiple RegressionAnalysisInternal validity:the statistical inferences about causal effects are valid forthe population being studied.Five threats to the internal validity of regression studies:Omitted variable biasWrong functional formErrors-in-variables biasSample selection biasSimultaneous causality bias
1.Omitted variable biasOmitted variable bias arises if an omitted variable isboth:
Instrumental VariableLets say you had two correlated variables that you wanted to regress: X and Y.Their correlation might be described by a third variable Z, which is associatedwith X in some way. Z is also associated with Y but only through Ys directassociation with XFor example, lets say you wanted to investigate the link between depression(X) and smoking (Y). Lack of job opportunities (Z) could lead to depression, butit is only associated with smoking through its association with depression (i.e.there isnt a direct correlation between lack of job opportunities and smoking).This third variable, Z (lack of job opportunities), can generally be used as aninstrumental variable if it can be measured and its behavior can be accountedfor
2.Wrong functional form (functional formmisspecification)Arises if the functional form is incorrectfor example, an interaction term isincorrectly omitted; then inferences on causal effects will be biased.Interaction: An interaction occurs when an independent variable has adifferent effect on the outcome depending on the values of anotherindependent variable.ŷ = b0+ b1X1+ b2X2ŷ = b0+ b1X1+ b2X2+ b3X1X2
3.Errors-in-variables biasIn reality, economic data often have measurement errorData entry errors in administrative dataRecollection errors in surveys (when did you start your current job?)Ambiguous questions (what was your income last year?)Intentionally false response problems with surveys (What is thecurrent value of your financial assets? How often do you drink anddrive?)
Solutions to errors-in-variables bias1.Obtain better data (often easier said than done).

Course Hero member to access this document