INR 3933 Exam 1 Study Guide
INR 3933 Exam 1 Study Guide INR3933
Popular in Special Topics in International Relations
Popular in Political Science
This 4 page Study Guide was uploaded by Michelle Alvarez on Monday September 28, 2015. The Study Guide belongs to INR3933 at Florida State University taught by Rob Carroll in Fall 2015. Since its upload, it has received 130 views. For similar materials see Special Topics in International Relations in Political Science at Florida State University.
Reviews for INR 3933 Exam 1 Study Guide
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/28/15
Exam 1 Recap Counterfactuals and Hypothesis Testing in Political Science Fearon What it is and why it s important If it had been the case that C or not C it would have been the case that E or not E Pg 169 What happens in the absence of the cause Counterfactuals make claims about events that did not actually occur pg 169 expresses what has not happened but could would or might under di ering conditions The closest possible world to the one we live in gt Example I took asprin for a headache so what would ve happened if I didn t take asprin X Necessary andor implicit in the efforts of political scientists to assess hypotheses about the causes of a phenomena being studied especially when there is a smallN because there are too many variables and too few cases any nonexperimental research that makes causal claims be it of the largeN or smallN variety must confront counterfactuals in the form of key assumptions or in the use of hypothetical comparison cases pg 194 Fearon s 3 main points of hypothesis testing I Counterfactual analysis counterfactuals are most likely to be found performing confirmatory work in case studies where the analyst is concerned with giving a causal explanation for some event or phenomenon causal claims may require arguments of counterfactual cases for justification pg 180 these arguments should be as explicit and defensible as possible 2 Experimental ideal Fearon suggests but does not advocate that counterfactuals play a key role in quasi or nonexperimental hypothesis testing that they do not play in actual experiments the comparison of actual cases vs counterfactual arguments pg 170171 gt Quasi or nonexperimental data not generated by random assignment to control and treatment groups 3 Manipulabilitv 0f the experiment variables are able to be manipulated to demonstrate causation 0 Statistics and Causal Inference Holland Causal inference drawing a conclusion about a causal connection based on the occurrence of an effect analyzes response of the effect variable when the cause is changed it is ultimately concerned with the effects of causes on specific units pg 947 Statistical reasoning can bring an emphasis on measuring the effects of causes rather than the causes of effects pg 945 Rubin s model for causal inference pg 946 gt Y is a response variable that measures the effect of the cause gt t is the treatment amp c is the control gt Yt u and Yc u are variables that represent two potential responses for a given unit u if the unit were exposed to t or c gt The fundamental problem of causal inference It is impossible to directly observe causal effects one cannot go back in history causation can only be inferred not exactly known You never get to see the same exact experiment under a control therefore you never get a real counterfactual It is impossible to observe the value of Yt u and Yc u on the same unit and therefore it is impossible to observe the effect of t on u pg 947 Example If the unit a is a specific fourth grader t represents a novel yearlong program of study of arithmetic c represents a standard arithmetic program and Y is a score on a test at the end of the year we could observe either Yt u or Yc u but not both pg 947 What constitutes a cause or potential cause pg 954955 gt A She did well on the exam because she is a woman The cause is an attribute that she possesses and cannot be a cause because potential exposability of the unit to all levels of the treatment is impossible We cannot manipulate attributes in this case gender or race The achievement score for a male would not be possible to conceive in A because gender is impossible to change on the same unitindividual gt B She did well on the exam because she studied for it The cause is a voluntary activity she performed It is not clear what would expose a person to studying or not Voluntary nature of human activity makes these types of causal statements difficult to apply to Rubin s model because the personal attribute decides if a treatment happens gt C She did well on the exam because she was coached by her teacher The cause was an actiVity that was imposed on her This is the same as an experimenter toggling with the treatment which easily matches Rubin s model and can be a cause In order for something to be a cause it has to be manipulatable gt NO CAUSATION WITHOUT MANIPULATION pg 959 Everything has a cause but it does not imply that everything can be a cause It is only impossible to make causal inference if assumptions are not tested 0 DISREGARD DAVID LEWIS READING
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'