LIR 832 – Lecture 8TuftePowerPoint PresentationClassic Approach to ResearchOpening of the Gender Earnings GapReview Literature and Develop a Theoretic ModelExplanations of Changes in Gender Differences in EarningsData and Empirical resultsSlide 9Slide 10Slide 11Slide 12Display of Regression ResultsImplicit Discussion of SpecificationThe Results: 1979Return to College EducationReturn to College QualityRepeat for 1986 estimates and compare to 1979ConclusionWhat You Need To Ask In Analyzing Statistical ResearchQuestions to Ask: The Empirical ModelQuestions to Ask: The SampleEmpirical MethodsQuestions: The ResultsResults: ContinuedQuestions: ConclusionsWorking with Qualitative DataDummy (Indicator) VariablesSlide 29Slide 30Slide 31Slide 32Slide 33Slide 34Slide 35Slide 36Slide 37Slide 38Slide 39Slide 40Slide 41Slide 42Slide 43Slide 44Slide 45Slide 46Slide 47Slide 48Slide 49InteractionsSlide 51Slide 52Slide 53Slide 54Slide 55Slide 56LIR 832 – Lecture 8The Good, the Bad and the Ugly:Tufte: Good and Bad Analysis of Quantative DataLoury: Gender Earnings GapQualitative DataTufteGroup 1: What did John Snow do in his analysis of the cholera epidemic that allowed him to support his theory (causal relation) about water and cholera?Group 2: What ended up obfuscating the relationship between temperature and O-ring failure in the analysis pre and post Challenger Disaster?Classic Approach to ResearchHow to write a research paper, an article and what to look for in statistical researchFind an interesting or pressing questionReview literature and develop a theoretic modelWhat is a theoretic model?b. Specify the empirical model: selection of independent variables and functional form.What do we mean by an empirical model? How is this different from a theoretic model?c. Hypothesize the expected signs of the coefficientsd. Collect the datae. Estimate and evaluate the equationf. Document the resultsOpening of the Gender Earnings GapWhat do we learn in the opening two paragraphs?i. The issue addressed by the paperii. What has gone beforeiii. How this paper contributes to the literature (knowledge)iv. Why is this important?Review Literature and Develop a Theoretic ModelRelevance of the Research QuestionCollege educated women made up a large and growing segment of the populationWhile most studies of the decline in the gender earnings gap present aggregate results across different years of schooling, factors that are responsible for narrowing the gender gap may differ by skill level.An analysis of the effects of college selectivity and academic performance makes it possible to determine whether gains have been uniform for all college educated women or whether some individuals have benefitted relative to others.Explanations of Changes in Gender Differences in Earningsrelative gain in the number of years worked by women compared to men.changed in perceptions about appropriate role of women in the labor marketchanges in industrial structure & the desired skill mix of workers within industryNote that each reason is developed and often supported by other informationData and Empirical resultsDescription of data used in the studyNLS data setslongitudinal (follow individuals over time)when and how collectedtypes of data collectedeasy to find the data and obtain additional information (standard data sets)Data is limited to white employees, why? How does this limit what we can learn from the study?What would be the natural next step?Table information:name in form readily understoodmean and standard deviation by genderdifference in means (what is missing: see fn10)focused discussion of factors most relevant to this study:male female contrasts between 1979 and 1986:raw weekly earningschanging gap in number of business and education majors by genderother germain factors:Display of Regression ResultsLay-Out of Table 2name, coefficient and SE of coefficient by gender and year.size of sample and r2Implicit Discussion of Specificationground on wage equations is very well plowed.education variables:college graduateyears of college 2 – 3College GPABusiness MajorEngineering MajorOther MajorCollege Characteristics:Median College Sat ScoreOther Characteristics:MarriedUnion MemberTenure on Current JobHow complete is this specification?The Results: 1979start by showing that the results conform with prior work (conventional wisdom) on factors which we are not immediately concerned withCut to the chase:returns to getting a degree are higher for women than for men:base is one year of college:estimate return for 2 - 3 years with no degreeestimate return to degreecompare return to degree to: those with one year of collegeReturn to College EducationCoefficient on College gradmen: .1214women .3173compare return to degree to those with 2 - 3 years of college (calculate significance of the difference:men: .1214 - .1004 = 0.02women .3173 - .1115 = 0.21show that this difference is statistically significantReturn to College QualityWomen’s earnings are more sensitive to the quality of the school than men’s earnings:Median SAT:men 0.0187 (.0214)women 0.0565 (.0282)Repeat for 1986 estimates and compare to 1979Return to graduating fell for women both absolutely and relative to menReturn to women’s GPA rose substantiallyLarger relative returns to Business major, larger returns to engineering major.Change in the structure of returns to college to women. In 1979 the return was to graduating from college, now it has more to do with performance in college and the major.Suggests that the changes in returns reflect within occupation changes in worker demand which favored women.ConclusionBrief summary of results and major points, doesn’t beat a dead horse.What You Need To Ask In Analyzing Statistical ResearchWhat is the question being asked by the researchers?How does this fit into prior research? Does the current research come out of prior research, or is it de novo with little base in previous work?De novo work requires closer examination than work which is firmly based in prior, established research.What is the ‘theoretic’ model is used by the researchers and what are the implications of this model? What hypotheses are derived about the relationship
View Full Document