Class 20Class OutlineSteps in the Elaboration ModelThree Scenarios of Elaboration AnalysisIntervening VariablesIntervening Variables: Berkeley Admission ExampleIntervening Variables: Berkeley Admission ExampleIntervening Variables: Euthanasia Attitude ExampleSlide 9Slide 10Confounding and Bias: Sex, Exercise, and Heart AttacksSlide 12Confounding and Bias: Age, Exercise, and Heart AttacksInteractionsSlide 15Slide 16Babbie’s Classification of Elaboration AnalysisSlide 18Class 20The Elaboration ModelClass Outline•The Elaboration Model—Analyses of Three Variables •Three Types of Elaboration Models–Intervening Variables–Confounding Variables–InteractionsSteps in the Elaboration Model1. A relationship is observed to exist between two variables.2. A third variable is held constant in the sense that the cases under study are subdivided according to the attributes of that third variable.3. The original two-variable relationship is recomputed within each of the subgroups.4. The comparison of the original relationship with the relationships found within each subgroup provides a fuller understanding of the original relationship itself.Three Scenarios of Elaboration AnalysisThe 3rd variable is an intervening variable.The 3rd variable is a confounding variable.The 3rd variable interacts with the independent variable.XYZXYZXYZ1.2.3.Intervening VariablesThree steps to show that Z explains the effect of independent variable X on dependent variable Y.1. Argue that X causally precedes Z, and Z causally precedes Y.2. Show that X and Z are correlated, Z and Y are correlated.3. Show that the correlation between X and Y is weakened after we control for Z in the model.Intervening Variables: Berkeley Admission ExamplegenderadmissionMale Femaleapplicants % admitted applicants % admitted2691 45% 1835 30%Gender affects admission rate.Intervening Variables:Berkeley Admission ExampleMale FemaleMajor applicants % admitted applicants % admittedA 825 62% 108 82%B 560 63% 25 68%C 325 37% 593 34%D 417 33% 375 35%E 191 28% 393 24%F 373 6% 341 7%total 2691 45% 1835 30%genderadmissionmajorAfter controlling for major, there is no NET effect of gender on admission rate.Intervening Variables:Euthanasia Attitude ExampleReligiosityPolitical viewslib=3, cons=1LibertarianismRaceblack=1Support for euthanasia++++--Note: Even though “political views” is an intervening variable, it does not explain blacks’ less support for euthanasia.Intervening Variables:Euthanasia Attitude ExampleReligiosityRaceblack=1+Intervening Variables:Euthanasia Attitude ExampleSupport for euthanasia-ReligiosityConfounding and Bias:Sex, Exercise, and Heart AttacksExerciseSex(Female=1)Heart attack---Leaving out the confounding variable sex in the analysis would underestimate the effect of exercise on the risk of heart attacks.Confounding and Bias:Sex, Exercise, and Heart AttacksTotal Female MaleSample # had a heart attackSample # had a heart attackSample # had a heart attackWeekly exercise300 30 100 6 200 24No weekly exercise300 48 200 24 100 24Note: Not real dataIf we don’t consider sex, weekly exercise reduces the risk of heart attach from 0.16 (48/300) to 0.1 (30/300). However, if we consider that men exercise more and are also more likely to get heart attacks, weekly exercise reduces the risk from 0.12 to 0.06 for women and from 0.24 to 0.12 for men.Confounding and Bias:Age, Exercise, and Heart AttacksExerciseAgeHeart attack--+Leaving out the confounding variable age in the analysis would overestimate the effect of exercise on the risk of heart attacks.Interactions•We say that X and Z have interactive effects on Y when –the effect of X on Y depends on the level of Z,–or equivalently, the effect of Z on Y depends on the level of X.•Graphically,XYZZYXIdentical phenomenonZ=1Z=0XYZ=1Z=0XYInteraction between X and Z No interaction between X and ZInteractionsPassiveActiveBelief in individual rightsInteractions•Example:–X = Libertarian attitude –Y = Support for euthanasia –Z = Type of euthanasia (passive or active)•Read the graph on the previous slide and choose the correct word for each statement below:–Libertarian attitude has a (greater/smaller) effect on support for passive euthanasia than that for active euthanasia.–The type of euthanasia makes a (greater/smaller) difference for individuals with libertarian views.Babbie’s Classification of Elaboration Analysis•Replication – A set of partial relationships is essentially the same as the corresponding zero-order relationship.•Explanation - A set of partial relationships is reduced essentially to zero when an antecedent variable is held constant.Babbie’s Classification of Elaboration Analysis•Interpretation - A set of partial relationships is reduced essentially to zero when an intervening variable is held constant).•Specification - One partial relationship is reduced, ideally to zero, and the other remains about the same as the original relationship or is
View Full Document